Search

With censored event time observations, the logrank test is the most popular tool for testing the equality of two underlying survival distributions. Although this test is asymptotically distribution free, it may not be powerful when the proportional hazards assumption is violated. Various other novel testing procedures have been proposed, which generally are derived by assuming a class of specific alternative hypotheses with respect to the hazard functions. The test considered by Pepe and Fleming (1989) is based on a linear combination of weighted differences of the two Kaplan-Meier curves over time and is a natural tool to assess the difference of two survival functions directly. In this article, we take a similar approach but choose weights that are proportional to the observed standardized difference of the estimated survival curves at each time point. The new proposal automatically makes weighting adjustments empirically. The new test statistic is aimed at a one-sided general alternative hypothesis and is distributed with a short right tail under the null hypothesis but with a heavy tail under the alternative. The results from extensive numerical studies demonstrate that the new procedure performs well under various general alternatives with a caution of a minor inflation of the type I error rate when the sample size is small or the number of observed events is small. The survival data from a recent cancer comparative study are utilized for illustrating the implementation of the process., (Copyright © 2015 John Wiley & Sons, Ltd.)

The propensity score is defined as a subject's probability of treatment selection, conditional on observed baseline covariates. Weighting subjects by the inverse probability of treatment received creates a synthetic sample in which treatment assignment is independent of measured baseline covariates. Inverse probability of treatment weighting (IPTW) using the propensity score allows one to obtain unbiased estimates of average treatment effects. However, these estimates are only valid if there are no residual systematic differences in observed baseline characteristics between treated and control subjects in the sample weighted by the estimated inverse probability of treatment. We report on a systematic literature review, in which we found that the use of IPTW has increased rapidly in recent years, but that in the most recent year, a majority of studies did not formally examine whether weighting balanced measured covariates between treatment groups. We then proceed to describe a suite of quantitative and qualitative methods that allow one to assess whether measured baseline covariates are balanced between treatment groups in the weighted sample. The quantitative methods use the weighted standardized difference to compare means, prevalences, higher-order moments, and interactions. The qualitative methods employ graphical methods to compare the distribution of continuous baseline covariates between treated and control subjects in the weighted sample. Finally, we illustrate the application of these methods in an empirical case study. We propose a formal set of balance diagnostics that contribute towards an evolving concept of 'best practice' when using IPTW to estimate causal treatment effects using observational data., (© 2015 The Authors. Statistics in Medicine Published by John Wiley & Sons Ltd.)

Introduction: Post-operative new-onset atrial fibrillation and flutter (POAF) is associated with increased morbidity and mortality following cardiac surgery. Registers and databases are important data sources for observational studies in this research area; hence, the aim was to assess the data validity of the POAF diagnosis in the Western Denmark Heart Registry (WDHR)., Methods: We studied a 25% random sample (n = 1,381) from a cohort of 5,532 patients who underwent coronary artery bypass grafting, valve surgery or combinations between 1 January 2011 and 31 December 2013. Registrations of POAF diagnoses in the WDHR were compared with the actual clinical course as documented in the medical records. The positive predictive value for new-onset POAF in the WDHR was calculated., Results: A positive predictive value of 82.5% (95% confidence interval: 78.8-85.7) was found with a sensitivity and specificity of 75.2% and 90.9%, respectively., Conclusions: A diagnosis of new-onset POAF in the WDHR is relatively valid and may be used for contemporary epidemiological studies. Improvements may optimise the registry's validity even further, emphasising the importance of continuous validation and maintenance of the registry., Funding: none., Trial Registration: not relevant.

Objectives: The propensity score (PS) is a balancing score. Following PS matching, balance checking usually relies on estimating separately the standardized absolute mean difference for each baseline characteristic. The average standardized absolute mean difference and the Mahalanobis distances have been proposed to summarize the information across the covariates. However, they might be minimized when nondesirable variables such as instrumental variables (IV) are included in the PS model. We propose a new weighted summary balance measure that takes into account, for each covariate, its strength of association with the outcome., Study Design and Setting: This new measure was evaluated using a simulation study to assess whether minimization of the measure coincided with minimally biased estimates. All measures were then applied to a real data set from an observational cohort study., Results: Contrarily to the other measures, our proposal was minimized when including the confounders, which coincided with minimal bias and mean squared error, but increased when including an IV in the PS model. Similar findings were observed in the real data set., Conclusion: A balance measure taking into account the strength of association between the covariates and the outcome may be helpful to identify the most parsimonious PS model., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Background: The prognosis of non-Hodgkin lymphoma and Hodgkin lymphoma is not affected by HIV serostatus, yet people living with HIV (PLWH) are frequently excluded from clinical trials in lymphoma., Methods: The UK NIHR Clinical Research Network Study Portfolio website was used to identify all the open clinical trials in lymphoma in the United Kingdom in January 2015. Trials that excluded PLWH were further investigated to evaluate if the exclusion was justified by scientific evidence., Results: We identified 56 multicentre open clinical trials in lymphoma including 46 interventional trials. People living with HIV were excluded from 32 interventional trials (70%). We identified a biologically valid reason (a potential increased risk of greater immunosuppression) for excluding PLWH from one trial and possibly for one optional arm in another study., Conclusions: There was no scientific or safety justification for excluding PLWH from most lymphoma clinical trials included on the NIHR portfolio. A clear justification for excluding PLWH was not offered in the available protocols. The exclusion of PLWH should be explicitly justified on scientific grounds in protocols to minimise stigmatisation.

Objective: To assess the degree of agreement between propensity score studies and randomized clinical trials in critical care research., Data Sources: Propensity score studies published in highly cited critical care or general medicine journals or included in a previous systematic review; corresponding randomized clinical trials included in Cochrane Systematic Reviews or published in PubMed., Study Selection: We identified propensity score studies of the effects of therapeutic interventions on short- or long-term mortality. We systematically matched propensity score studies to randomized clinical trials based on patient selection criteria, interventions, and outcomes., Data Extraction: We appraised the methods of included studies and extracted treatment effect estimates to compare the results of propensity score studies and randomized clinical trials. When multiple studies were identified for the same topic, we performed meta-analyses to obtain summary treatment effect estimates., Data Synthesis: We matched 21 propensity score studies with 58 randomized clinical trials in 18 distinct comparisons (median, one propensity score study and two randomized clinical trials per comparison), for short- and long-term mortality. We found one statistically significant difference between designs (hyperoncotic albumin vs crystalloid fluids) among these 18 comparisons. Propensity score studies did not produce systematically higher (or lower) treatment effect estimates compared with randomized clinical trials, but estimates from the two designs differed by more than 30% in one third of the comparisons examined. Observational studies in critical care met widely accepted methodological standards for propensity score analyses., Conclusions: Across diverse critical care topics, propensity score studies published in high-impact journals produced results that were generally consistent with the findings of randomized clinical trials. However, caution is needed when interpreting propensity score studies because occasionally their results contradict those of randomized clinical trials and there is no reliable way to predict disagreements.

We propose a simple approach predicting the cumulative risk of disease accommodating predictors with time-varying effects and outcomes subject to censoring. We use a nonparametric function for the coefficient of the time-varying effect and handle censoring through self-consistency equations that redistribute the probability mass of censored outcomes to the right. The computational procedure is extremely convenient and can be implemented by standard software. We prove large sample properties of the proposed estimator and evaluate its finite sample performance through simulation studies. We apply the method to estimate the cumulative risk of developing Huntington's disease (HD) from subjects with huntingtin gene mutation using a large collaborative HD study data and illustrate an inverse relationship between the cumulative risk of HD and the length of cytosine-adenine-guanine repeats in the huntingtin gene., (Copyright © 2015 John Wiley & Sons, Ltd.)

Aims: The observational PICTURE (Place of Reveal In the Care pathway and Treatment of patients with Unexplained Recurrent Syncope) registry enrolled 570 patients with unexplained syncope, documented their care pathway and the various tests they underwent before the insertion of an implantable loop recorder (ILR). The aims were to describe the extent and cost of diagnostic tests performed before the implant., Methods and Results: Actual costs of 17 predefined diagnostic tests were characterized based on a combination of data from PICTURE and a micro-costing study performed at a medium-sized UK university hospital in the UK. The median cost of diagnostic tests per patient was £1114 (95% CI £995-£1233). As many patients received more than the median number of tests, the mean expenditure per patient was higher with £1613 (95% CI £1494-£1732), and for 10% of the patients the cost exceeded £3539. Tests were frequently repeated, and early use of specific and expensive tests was common. In the 12% of patients with types of tests entirely within the recommendations for an initial evaluation before ILR implant, the mean cost was £710., Conclusion: Important opportunities to reduce test-related costs before an ILR implant were identified, e.g. by more appropriate use of tests recommended in the initial evaluation, by decreasing repetition of tests, and by avoiding early use of specialized and expensive tests. A structured multidisciplinary approach would be the best model to achieve an optimal outcome., (© The Author 2015. Published by Oxford University Press on behalf of the European Society of Cardiology.)

Purpose: The generalisability of randomised controlled trials (RCTs) may be limited by restrictive entry criteria or by their experimental nature. Observational research can provide complementary findings but is prone to bias. Employing propensity score matching, to reduce such bias, we compared the real-life effect of cinacalcet use on all-cause mortality (ACM) with findings from the Evaluation of Cinacalcet Therapy to Lower Cardiovascular Events (EVOLVE) RCT in chronic haemodialysis patients., Methods: Incident adult haemodialysis patients receiving cinacalcet, recruited in a prospective observational cohort from 2007-2009 (AROii; n = 10,488), were matched to non-exposed patients regardless of future exposure status. The effect of treatment crossover was investigated with inverse probability of censoring weighted and lag-censored analyses. EVOLVE ACM data were analysed largely as described for the primary composite endpoint., Results: AROii patients receiving cinacalcet (n = 532) were matched to 1790 non-exposed patients. The treatment effect of cinacalcet on ACM in the main AROii analysis (hazard ratio 1.03 [95% confidence interval (CI) 0.78-1.35]) was closer to the null than for the Intention to Treat (ITT) analysis of EVOLVE (0.94 [95%CI 0.85-1.04]). Adjusting for non-persistence by 0- and 6-month lag-censoring and by inverse probability of censoring weight, the hazard ratios in AROii (0.76 [95%CI 0.51-1.15], 0.84 [95%CI 0.60-1.18] and 0.79 [95%CI 0.56-1.11], respectively) were comparable with those of EVOLVE (0.82 [95%CI 0.67-1.01], 0.83 [95%CI 0.73-0.96] and 0.87 [95%CI 0.71-1.06], respectively)., Conclusions: Correcting for treatment crossover, we observed results in the 'real-life' setting of the AROii observational cohort that closely mirrored the results of the EVOLVE RCT. Persistence-corrected analyses revealed a trend towards reduced ACM in haemodialysis patients receiving cinacalcet therapy., (Copyright © 2015 John Wiley & Sons, Ltd.)

Objective: To explore contributing factors (CF) associated to related critical patients safety incidents., Design: SYREC study pos hoc analysis., Setting: A total of 79 Intensive Care Departments were involved., Patients: The study sample consisted of 1.017 patients; 591 were affected by one or more incidents., Main Variables: The CF were categorized according to a proposed model by the National Patient Safety Agency from United Kingdom that was modified. Type, class and severity of the incidents was analyzed., Results: A total 2,965 CF were reported (1,729 were associated to near miss and 1,236 to adverse events). The CF group more frequently reported were related patients factors. Individual factors were reported more frequently in near miss and task related CF in adverse events. CF were reported in all classes of incidents. The majority of CF were reported in the incidents classified such as less serious, even thought CF patients factors were associated to serious incidents. Individual factors were considered like avoidable and patients factors as unavoidable., Conclusions: The CF group more frequently reported were patient factors and was associated to more severe and unavoidable incidents. By contrast, individual factors were associated to less severe and avoidable incidents. In general, CF most frequently reported were associated to near miss., (Copyright © 2014 Elsevier España, S.L.U. and SEMICYUC. All rights reserved.)

Background: The aim of this meta-analysis was to investigate possible relationships between bariatric surgery and incidence of obesity-related cancers. Obesity is an established risk factor for obesity-related cancers but the effects of bariatric surgery on incidence of obesity-related cancers are uncertain., Material/methods: We searched 4 electronic databases to identify eligible studies: PubMed, Embase, Web of Science, and Google Scholar. Five observational studies were eligible and included in this meta-analysis. Random-effects or fixed-effects odds ratio (OR) and its corresponding 95% confidence interval (CI) were pooled., Results: Meta-analysis of these 5 observational studies revealed that bariatric surgery was associated with a significantly (p=0.0004) reduced incidence of obesity-related cancers (OR=0.43, 95%CI, 0.27-0.69) when compared with control individuals. Pooled estimated data showed that bariatric surgery is associated with a 24% lower colorectal cancer (CRC) risk. No publication bias was detected by Egger's or Begg's tests., Conclusions: Although bariatric surgery may significantly reduce incidence of obesity-related cancers, considering the limitations of these included studies, these findings should be confirmed by further well-designed studies.

Background: Whether self-reported daytime napping is an independent predictor of cardiovascular or all-cause mortality remains unclear. The aim of this study was to investigate self-reported daytime napping and risk of cardiovascular or all-cause mortality by conducting a meta-analysis., Material/methods: A computerized literature search of PubMed, Embase, and Cochrane Library was conducted up to May 2014. Only prospective studies reporting risk ratio (RR) and corresponding 95% confidence intervals (CI) of cardiovascular or all-cause mortality with respect to baseline self-reported daytime napping were included., Results: Seven studies with 98,163 subjects were included. Self-reported daytime napping was associated with a greater risk of all-cause mortality (RR 1.15; 95% CI 1.07-1.24) compared with non-nappers. Risk of all-cause mortality appeared to be more pronounced among persons with nap duration >60 min (RR 1.15; 95% CI 1.04-1.27) than persons with nap duration <60 min (RR 1.10; 95% CI 0.92-1.32). The pooled RR of cardiovascular mortality was 1.19 (95% CI 0.97-1.48) comparing daytime nappers to non-nappers., Conclusions: Self-reported daytime napping is a mild but statistically significant predictor for all-cause mortality, but not for cardiovascular mortality. However, whether the risk is attributable to excessive sleep duration or napping alone remains controversial. More prospective studies stratified by sleep duration, napping periods, or age are needed.

Background: Some epidemiological studies have suggested that vitamin E intake reduces the risk of pancreatic cancer; however, this conclusion has not been supported by all the published studies. We conducted a meta-analysis to assess the relationship between vitamin E intake and the risk of pancreatic cancer by combining the results from published articles., Material/methods: We searched the published studies that reported the relationship between vitamin E intake and pancreatic cancer risk using the PubMed, Web of Science, and Embase databases through December 31st, 2014. Based on a fixed-effects or random-effects model, the RR and 95% CI were used to assess the combined risk., Results: In total, 10 observational studies (6 case-control studies and 4 cohort studies) were included. The overall RR (95% CI) of pancreatic cancer for the highest vs. the lowest level of vitamin E intake was 0.81 (0.73, 0.89). We found little evidence of heterogeneity (I2=19.8%, P=0.255). In the subgroup analyses, we found an inverse association between vitamin E intake and pancreatic cancer risk both in the case-control and cohort studies. Additionally, this inverse association was not modified by different populations., Conclusions: In our meta-analysis, there was an inverse association between vitamin E intake and the risk of pancreatic cancer. A high level of vitamin E might be a protective factor for populations at risk for pancreatic cancer.

Introduction: In critical care observational studies, when clinicians administer different treatments to sicker patients, any treatment comparisons will be confounded by differences in severity of illness between patients. We sought to investigate the extent that observational studies assessing treatments are at risk of incorrectly concluding such treatments are ineffective or even harmful due to inadequate risk adjustment., Methods: We performed Monte Carlo simulations of observational studies evaluating the effect of a hypothetical treatment on mortality in critically ill patients. We set the treatment to have either no association with mortality or to have a truly beneficial effect, but more often administered to sicker patients. We varied the strength of the treatment's true effect, strength of confounding, study size, patient population, and accuracy of the severity of illness risk-adjustment (area under the receiver operator characteristics curve, AUROC). We measured rates in which studies made inaccurate conclusions about the treatment's true effect due to confounding, and the measured odds ratios for mortality for such false associations., Results: Simulated observational studies employing adequate risk-adjustment were generally able to measure a treatment's true effect. As risk-adjustment worsened, rates of studies incorrectly concluding the treatment provided no benefit or harm increased, especially when sample size was large (n = 10,000). Even in scenarios of only low confounding, studies using the lower accuracy risk-adjustors (AUROC < 0.66) falsely concluded that a beneficial treatment was harmful. Measured odds ratios for mortality of 1.4 or higher were possible when the treatment's true beneficial effect was an odds ratio for mortality of 0.6 or 0.8., Conclusions: Large observational studies confounded by severity of illness have a high likelihood of obtaining incorrect results even after employing conventionally "acceptable" levels of risk-adjustment, with large effect sizes that may be construed as true associations. Reporting the AUROC of the risk-adjustment used in the analysis may facilitate an evaluation of a study's risk for confounding.

Statistical tests in medical research: traditional methods vs. multivariate npc permutation tests.Within medical research, a useful statistical tool is based on hypotheses testing in terms of the so-called null, that is the treatment has no effect, and alternative hypotheses, that is the treatment has some effects. By controlling the risks of wrong decisions, empirical data are used in order to possibly reject the null hypotheses in favour of the alternative, so that demonstrating the efficacy of a treatment of interest. The multivariate permutation tests, based on the nonparametric combination - NPC method, provide an innovative, robust and effective hypotheses testing solution to many real problems that are commonly encountered in medical research when multiple end-points are observed. This paper discusses the various approaches to hypothesis testing and the main advantages of NPC tests, which consist in the fact that they require much less stringent assumptions than traditional statistical tests. Moreover, the related results may be extended to the reference population even in case of selection-bias, that is non-random sampling. In this work, we review and discuss some basic testing procedures along with the theoretical and practical relevance of NPC tests showing their effectiveness in medical research. Within the non-parametric methods, NPC tests represent the current "frontier" of statistical research, but already widely available in the practice of analysis of clinical data.

Context: Several previous observational studies showed an association between hypopituitarism and excess mortality. Reports on reduction of standard mortality ratio (SMR) with GH replacement have been published recently., Objective: This meta-analysis assessed studies reporting SMR to clarify mortality risk in hypopituitary adults and also the potential benefit conferred by GH replacement., Data Sources: A literature search was performed in Medline, Embase, and Cochrane library up to March 31, 2014., Eligibility Criteria: Studies with or without GH replacement reporting SMR with 95% confidence intervals (95% CI) were included., Data Extraction and Analysis: Patient characteristics, SMR data, and treatment outcomes were independently assessed by two authors, and with consensus from third author, studies were selected for analysis. Meta-analysis was performed in all studies together, and those without and with GH replacement separately, using the statistical package metafor in R., Results: Six studies reporting a total of 19 153 hypopituiatary adults with a follow-up duration of more than 99,000 person years were analyzed. Hypopituitarism was associated with an overall excess mortality (weighted SMR, 1.99; 95% CI, 1.21-2.76) in adults. Female hypopituitary adults showed higher SMR compared with males (2.53 vs 1.71). Onset of hypopituitarism at a younger age was associated with higher SMR. GH replacement improved the mortality risk in hypopituitary adults that is comparable to the background population (SMR with GH replacement, 1.15; 95% CI, 1.05-1.24 vs SMR without GH, 2.40; 95% CI, 1.46-3.34). GH replacement conferred lower mortality benefit in hypopituitary women compared with men (SMR, 1.57; 95% CI, 1.38-1.77 vs 0.95; 95% CI, 0.85-1.06)., Limitations: There was a potential selection bias of benefit of GH replacement from a post-marketing data necessitating further evidence from long-term randomized controlled trials., Conclusions: Hypopituitarism may increase premature mortality in adults. Mortality benefit from GH replacement in hypopituitarism is less pronounced in women than men.

Post-licensure medical product safety surveillance is important for detecting adverse events potentially not identified pre-licensure. Historically, post-licensure safety monitoring has been accomplished using passive reporting systems and by conducting formal Phase IV randomized trials or large epidemiological studies, also known as safety surveillance or pharmacovigilance studies. However, crucial gaps in the safety evidence base provided by these approaches have led to high profile product withdrawals and growing public concern about unknown health risks associated with licensed products. To address the limitations of existing surveillance systems and to facilitate more accurate and rapid detection of safety problems, new systems involving active surveillance of large, population-based cohorts using observational health care databases are being developed. In this article, we review common statistical methods that have been employed previously for post-licensure safety monitoring, including data mining and sequential hypothesis testing, and assess which methods may be promising for potential use within this newly proposed prospective observational cohort monitoring framework. We discuss gaps in existing approaches and identify areas where methodological development is needed to improve the success of safety surveillance efforts in this setting., (© The Author(s) 2011 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav.)

Sequential methods are well established for randomized clinical trials (RCTs), and their use in observational settings has increased with the development of national vaccine and drug safety surveillance systems that monitor large healthcare databases. Observational safety monitoring requires that sequential testing methods be better equipped to incorporate confounder adjustment and accommodate rare adverse events. New methods designed specifically for observational surveillance include a group sequential likelihood ratio test that uses exposure matching and generalized estimating equations approach that involves regression adjustment. However, little is known about the statistical performance of these methods or how they compare to RCT methods in both observational and rare outcome settings. We conducted a simulation study to determine the type I error, power and time-to-surveillance-end of group sequential likelihood ratio test, generalized estimating equations and RCT methods that construct group sequential Lan-DeMets boundaries using data from a matched (group sequential Lan-DeMets-matching) or unmatched regression (group sequential Lan-DeMets-regression) setting. We also compared the methods using data from a multisite vaccine safety study. All methods had acceptable type I error, but regression methods were more powerful, faster at detecting true safety signals and less prone to implementation difficulties with rare events than exposure matching methods. Method performance also depended on the distribution of information and extent of confounding by site. Our results suggest that choice of sequential method, especially the confounder control strategy, is critical in rare event observational settings. These findings provide guidance for choosing methods in this context and, in particular, suggest caution when conducting exposure matching., (Copyright © 2014 John Wiley & Sons, Ltd.)

Treatment preferences of groups (e.g., clinical centers) have often been proposed as instruments to control for unmeasured confounding-by-indication in instrumental variable (IV) analyses. However, formal evaluations of these group-preference-based instruments are lacking. Unique challenges include the following: (i) correlations between outcomes within groups; (ii) the multi-value nature of the instruments; (iii) unmeasured confounding occurring between and within groups. We introduce the framework of between-group and within-group confounding to assess assumptions required for the group-preference-based IV analyses. Our work illustrates that, when unmeasured confounding effects exist only within groups but not between groups, preference-based IVs can satisfy assumptions required for valid instruments. We then derive a closed-form expression of asymptotic bias of the two-stage generalized ordinary least squares estimator when the IVs are valid. Simulations demonstrate that the asymptotic bias formula approximates bias in finite samples quite well, particularly when the number of groups is moderate to large. The bias formula shows that when the cluster size is finite, the IV estimator is asymptotically biased; only when both the number of groups and cluster size go to infinity, the bias disappears. However, the IV estimator remains advantageous in reducing bias from confounding-by-indication. The bias assessment provides practical guidance for preference-based IV analyses. To increase their performance, one should adjust for as many measured confounders as possible, consider groups that have the most random variation in treatment assignment and increase cluster size. To minimize the likelihood for these IVs to be invalid, one should minimize unmeasured between-group confounding., (Copyright © 2014 John Wiley & Sons, Ltd.)

Purpose: Evidence-based medicine requires primary results from randomized controlled trials and high-quality observational studies; however, in pediatric medicine it can be difficult to enroll sufficient numbers of children to reach sample sizes required for statistical significance. The experience of clinicians and researchers with the practicalities of recruitment has not been explored, but could potentially inform best practice and lead to the development of successful research networks. Perceived barriers and facilitators to recruitment in child eye health in the United Kingdom, where adoption of trials and studies onto a central database has created a register of high quality projects, were investigated., Methods: Studies and trials in child eye health from the United Kingdom Clinical Research Network portfolio database were identified and the named researchers contacted. Based on a validated tool, an electronic survey was created to ask about recruitment experience to identify facilitators and barriers., Results: There were 46 trials and studies recruiting children, and 51 completed questionnaires from researchers were received. Children's eye research activity is mainly based at tertiary referral centers. Although most studies recruit to target, many researchers report difficulties. Availability of dedicated research teams in clinics and good communication between research team, clinical team, and families are strong facilitators. Administrative difficulties and the burden of research to clinical teams and families are strong barriers., Conclusions: Support for investigators in the form of dedicated research and administrative staff and good communication enhance recruitment to studies in child eye health., (Copyright 2015, SLACK Incorporated.)

Aims: Toll-like receptor 4 (TLR4) gene Asp299Gly and Thr399Ile polymorphisms have been reported to be associated with susceptibility to type 2 diabetes mellitus (T2DM) with inconsistent results. In an effort to clarify earlier inconclusive results, a meta-analysis evaluating the precise associations between TLR4 gene Asp299Gly and Thr399Ile polymorphisms and T2DM risk was performed., Methods: We searched the PubMed, Embase, Web of Science, CBMdisc, CNKI, and Google Scholar until July 17, 2014. Additionally, hand searching of the references of identified articles were performed. Original observational studies dealing with the associations between TLR4 gene Asp299Gly and Thr399Ile polymorphisms and T2DM risk were selected. Heterogeneity and publication bias were determined and the meta-analysis was performed by Review Manager 5.1.4 and Stata 11.0., Results: Seventeen articles involving 25 studies were included in the final meta-analysis, covering a total of 5963 T2DM cases and 9096 controls. For TLR4 gene Asp299Gly polymorphism, 17 studies were combined showing no evidence for association between TLR4 gene Asp299Gly polymorphism and T2DM risk. For TLR4 gene Thr399Ile polymorphism, eight studies were combined. There was also lack of evidence for significant association between TLR4 gene Thr399Ile polymorphism and T2DM risk. In addition, the similar results were obtained in the sensitivity analyses., Conclusions: The present meta-analysis indicates that TLR4 gene Asp299Gly and Thr399Ile polymorphisms are not associated with increased T2DM risk., (Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.)

Background: Experimental and clinical studies have suggested that the presence of fragmented QRS complex (fQRS) is associated with various cardiovascular diseases. fQRS may predict major adverse cardiovascular events (MACE). The current meta-analysis was performed using clinical outcome studies to evaluate the role of fQRS in coronary artery disease (CAD)., Methods: A systematic search of electronic databases (Cochrane, Medline, Embase and Pubmed) from their inception to April 2014 was performed. Data were extracted from applicable articles to evaluate the prognostic value of fQRS in CAD., Results: A total of 16 observational studies about fQRS and CAD (n = 3,997 patients) were identified. Compared with the non-fQRS group, MACE and mortality were significantly higher in the fQRS group -odds ratios (OR) 3.19, 95 % confidence interval (95 % CI) [2.3, 4.42], p < 0.00001; OR 2.24, 95 % CI [1.71, 2.94], p < 0.0001. Patients developed Q waves, anterior-wall myocardial infarction (MI), and low left ventricular ejection fraction (LVEF) more frequently in the fQRS group than in the non-fQRS group-OR 2.59, 95 % CI [1.76, 3.81], p < 0.00001; OR 2.43, 95 % CI [1.07, 5.52], p = 0.03; OR - 6.43, 95 % CI [- 9.11, - 3.74], p < 0.00001., Conclusion: Based on current evidence, fQRS was associated with increased MACE, mortality, Q waves, anterior-wall MI, and decreased LVEF in CAD. These findings show that fQRS is a reliable marker in CAD.

Marginal structural models are commonly used to estimate the causal effect of a time-varying treatment in presence of time-dependent confounding. When fitting an MSM to data, the analyst must specify both the structural model for the outcome and the treatment models for the inverse-probability-of-treatment weights. The use of stabilized weights is recommended because they are generally less variable than the standard weights. In this paper, we are concerned with the use of the common stabilized weights when the structural model is specified to only consider partial treatment history, such as the current or most recent treatments. We present various examples of settings where these stabilized weights yield biased inferences while the standard weights do not. These issues are first investigated on the basis of simulated data and subsequently exemplified using data from the Honolulu Heart Program. Unlike common stabilized weights, we find that basic stabilized weights offer some protection against bias in structural models designed to estimate current or most recent treatment effects., (Copyright © 2014 John Wiley & Sons, Ltd.)

The high-dimensional propensity score (hdPS) algorithm was proposed for automation of confounding adjustment in problems involving large healthcare databases. It has been evaluated in comparative effectiveness research (CER) with point treatments to handle baseline confounding through matching or covariance adjustment on the hdPS. In observational studies with time-varying interventions, such hdPS approaches are often inadequate to handle time-dependent confounding and selection bias. Inverse probability weighting (IPW) estimation to fit marginal structural models can adequately handle these biases under the fundamental assumption of no unmeasured confounders. Upholding of this assumption relies on the selection of an adequate set of covariates for bias adjustment. We describe the application and performance of the hdPS algorithm to improve covariate selection in CER with time-varying interventions based on IPW estimation and explore stabilization of the resulting estimates using Super Learning. The evaluation is based on both the analysis of electronic health records data in a real-world CER study of adults with type 2 diabetes and a simulation study. This report (i) establishes the feasibility of IPW estimation with the hdPS algorithm based on large electronic health records databases, (ii) demonstrates little impact on inferences when supplementing the set of expert-selected covariates using the hdPS algorithm in a setting with extensive background knowledge, (iii) supports the application of the hdPS algorithm in discovery settings with little background knowledge or limited data availability, and (iv) motivates the application of Super Learning to stabilize effect estimates based on the hdPS algorithm., (Copyright © 2014 John Wiley & Sons, Ltd.)

Instrumental variable (IV) analysis has been widely used in economics, epidemiology, and other fields to estimate the causal effects of covariates on outcomes, in the presence of unobserved confounders and/or measurement errors in covariates. However, IV methods for time-to-event outcome with censored data remain underdeveloped. This paper proposes a Bayesian approach for IV analysis with censored time-to-event outcome by using a two-stage linear model. A Markov chain Monte Carlo sampling method is developed for parameter estimation for both normal and non-normal linear models with elliptically contoured error distributions. The performance of our method is examined by simulation studies. Our method largely reduces bias and greatly improves coverage probability of the estimated causal effect, compared with the method that ignores the unobserved confounders and measurement errors. We illustrate our method on the Women's Health Initiative Observational Study and the Atherosclerosis Risk in Communities Study., (Copyright © 2014 John Wiley & Sons, Ltd.)

Inhaled corticosteroids are commonly prescribed for patients with severe chronic obstructive pulmonary disease. Although their use improves quality of life and reduces exacerbations, it is associated with increased risk of pneumonia. Curiously, their use has not been associated with increased risk of pneumonia-related or overall mortality. We review pertinent literature to further explore the effects of inhaled corticosteroids on incident pneumonia and mortality in patients with chronic obstructive pulmonary disease. The association of use of inhaled corticosteroids and incident pneumonia is substantial and has been present in the majority of the studies on the topic. This includes both randomized controlled trials and observational studies. However, all of the studies have substantial risk of bias. Most randomized trials are limited by lack of systematic ascertainment of pneumonia; they depended on adverse event reporting. Many observational studies included proper radiographic assessment of pneumonia, but they are limited by their retrospective, observational design. The unadjusted higher risk of pneumonia is associated with longer duration of use, more potent ICS compounds, and higher doses. That implies a dose-effect relationship. Unlike pneumonia, mortality is a precise outcome. Despite the robust association of inhaled corticosteroid use with increased risk of pneumonia, all studies find either no difference or a reduction in pulmonary-related and overall mortality associated with the use of inhaled corticosteroids. These observations suggest a double effect of inhaled corticosteroids (i.e., an adverse effect plus an unexplained mitigating effect).

Aims: Near-infrared spectroscopy has the potential for aiding the diagnosis of major depressive disorder. The purpose of this study was to systematically review the evidence from observational studies regarding the use of near-infrared spectroscopy in patients with major depressive disorder and to identify the characteristic pattern of prefrontal lobe activity in major depressive disorder., Methods: medline, PubMed, Cochrane Library and Web of Science databases were searched in December 2013. All case-control studies were included. The quality of evidence was examined using the Newcastle-Ottawa Quality Assessment Scale. The primary outcome measures were the mean oxygenated and deoxygenated hemoglobin alterations of the cerebral cortex during cognitive activation periods. The standard mean difference for the overall pooled effects across the included studies was estimated using random or fixed effect models. The primary outcome measures were included in the meta-analysis., Results: Fourteen studies met the inclusion criteria. Six studies (n = 692 participants) were included in the analysis of the mean oxygenated hemoglobin alterations; the pooled mean standardized difference was -0.74 (95% confidence interval, -0.97 to -0.52), indicating that patients with major depressive disorder were associated with attenuated increase in oxygenated hemoglobin during cognitive activation in the prefrontal regions compared to healthy controls. Five studies (n = 668 participants) were included in the analysis of mean deoxygenated-hemoglobin changes; the pooled standardized mean difference was 0.18 (95% confidence interval, -0.20 to 0.56)., Conclusions: Using near-infrared spectroscopy measurements, we observed that compared to healthy subjects, patients with major depressive disorder had significantly lower prefrontal activation during cognitive tasks., (© 2014 The Authors. Psychiatry and Clinical Neurosciences © 2014 Japanese Society of Psychiatry and Neurology.)

We reviewed recent scientific evidence regarding the effects of multivitamin/mineral (MVM) supplements on risk of chronic diseases, including cancer, cardiovascular disease, and age-related eye diseases. Data from randomized controlled trials (RCTs) and observational, prospective cohort studies were examined. The majority of scientific studies investigating the use of MVM supplements in chronic disease risk reduction reported no significant effect. However, the largest and longest RCT of MVM supplements conducted to date, the Physicians' Health Study II (PHS II), found a modest and significant reduction in total and epithelial cancer incidence in male physicians, consistent with the Supplémentation en Vitamines et Minéraux Antioxydants (SU.VI.MAX) trial. In addition, PHS II found a modest and significant reduction in the incidence of nuclear cataract, in agreement with several other RCTs and observational, prospective cohort studies. The effects of MVM use on other subtypes of cataract and age-related macular degeneration remain unclear. Neither RCTs nor prospective cohort studies are without their limitations. The placebo-controlled trial design of RCTs may be inadequate for nutrient interventions, and residual confounding, measurement error, and the possibility of reverse causality are inherent to any observational study. National surveys show that micronutrient inadequacies are widespread in the US and that dietary supplements, of which MVMs are the most common type, help fulfill micronutrient requirements in adults and children.

The objective of this study was to perform a systematic review of published literature on differentially expressed genes (DEGs) in human epicardial adipose tissue (EAT) to identify molecules associated with CVDs. A systematic literature search was conducted in PubMed, SCOPUS, and ISI Web of Science literature databases for papers published before October 2014 that addressed EAT genes and cardiovascular diseases (CVDs). We included original papers that had performed gene expressions in EAT of patients undergoing open-heart surgery. The Reporting Recommendations for Tumor Marker Prognostic Studies (PRIMARK) assessment tool was also used for methodological quality assessment. From the 180 papers identified by our initial search strategy, 40 studies met the inclusion criteria and presented DEGs in EAT samples from patients with and without CVDs. The included studies reported 42 DEGs identified through comparison of EAT-specific gene expression in patients with and without CVDs. Among the 42 DEGs, genes involved in regulating apoptosis had higher enrichment scores. Notably, interleukin-6 (IL-6) and tumor protein p53 (TP53) were the main hub genes in the network. The results suggest that regulation of apoptosis in EAT is critical for CVD development. Moreover, IL-6 and TP53 as hub genes could serve as biomarkers and therapeutic targets for CVDs.

Background: Although randomization is considered essential for causal inference, it is often not possible to randomize in nutrition and obesity research. To address this, we develop a framework for an experimental design-packet randomized experiments (PREs), which improves causal inferences when randomization on a single treatment variable is not possible. This situation arises when subjects are randomly assigned to a condition (such as a new roommate) which varies in one characteristic of interest (such as weight), but also varies across many others. There has been no general discussion of this experimental design, including its strengths, limitations, and statistical properties. As such, researchers are left to develop and apply PREs on an ad hoc basis, limiting its potential to improve causal inferences among nutrition and obesity researchers., Methods: We introduce PREs as an intermediary design between randomized controlled trials and observational studies. We review previous research that used the PRE design and describe its application in obesity-related research, including random roommate assignments, heterochronic parabiosis, and the quasi-random assignment of subjects to geographic areas. We then provide a statistical framework to control for potential packet-level confounders not accounted for by randomization., Results: Packet randomized experiments have successfully been used to improve causal estimates of the effect of roommates, altitude, and breastfeeding on weight outcomes. When certain assumptions are met, PREs can asymptotically control for packet-level characteristics. This has the potential to statistically estimate the effect of a single treatment even when randomization to a single treatment did not occur., Conclusions: Applying PREs to obesity-related research will improve decisions about clinical, public health, and policy actions insofar as it offers researchers new insight into cause and effect relationships among variables., (© 2014 Stichting European Society for Clinical Investigation Journal Foundation.)

Objective: To pool data from completed amyotrophic lateral sclerosis (ALS) clinical trials and create an open-access resource that enables greater understanding of the phenotype and biology of ALS., Methods: Clinical trials data were pooled from 16 completed phase II/III ALS clinical trials and one observational study. Over 8 million de-identified longitudinally collected data points from over 8,600 individuals with ALS were standardized across trials and merged to create the Pooled Resource Open-Access ALS Clinical Trials (PRO-ACT) database. This database includes demographics, family histories, and longitudinal clinical and laboratory data. Mixed effects models were used to describe the rate of disease progression measured by the Revised ALS Functional Rating Scale (ALSFRS-R) and vital capacity (VC). Cox regression models were used to describe survival data. Implementing Bonferroni correction, the critical p value for 15 different tests was p = 0.003., Results: The ALSFRS-R rate of decline was 1.02 (±2.3) points per month and the VC rate of decline was 2.24% of predicted (±6.9) per month. Higher levels of uric acid at trial entry were predictive of a slower drop in ALSFRS-R (p = 0.01) and VC (p < 0.0001), and longer survival (p = 0.02). Higher levels of creatinine at baseline were predictive of a slower drop in ALSFRS-R (p = 0.01) and VC (p < 0.0001), and longer survival (p = 0.01). Finally, higher body mass index (BMI) at baseline was associated with longer survival (p < 0.0001)., Conclusion: The PRO-ACT database is the largest publicly available repository of merged ALS clinical trials data. We report that baseline levels of creatinine and uric acid, as well as baseline BMI, are strong predictors of disease progression and survival., (© 2014 American Academy of Neurology.)

Background: The role of observational studies in informing clinical practice is debated, and high profile examples of discrepancies between the results of observational studies and randomised controlled trials (RCTs) have intensified that debate. We systematically reviewed findings from the Nurses' Health Study (NHS), one of the longest and largest observational studies, to assess the number and strength of the associations reported and to determine if they have been confirmed in RCTs., Methods: We reviewed NHS publication abstracts from 1978-2012, extracted information on associations tested, and graded the strength of the reported effect sizes. We searched PubMed for RCTs or systematic reviews for 3 health outcomes commonly reported in NHS publications: breast cancer, ischaemic heart disease (IHD) and osteoporosis. NHS results were compared with RCT results and deemed concordant when the difference in effect sizes between studies was ≤0.15., Findings: 2007 associations between health outcomes and independent variables were reported in 1053 abstracts. 58.0% (1165/2007) were statistically significant, and 22.2% (445/2007) were neutral (no association). Among the statistically significant results that reported a numeric odds ratio (OR) or relative risk (RR), 70.5% (706/1002) reported a weak association (OR/RR 0.5-2.0), 24.5% (246/1002) a moderate association (OR/RR 0.25-0.5 or 2.0-4.0) and 5.0% (50/1002) a strong association (OR/RR ≤0.25 or ≥4.0). 19 associations reported in NHS publications for breast cancer, IHD and osteoporosis have been tested in RCTs, and the concordance between NHS and RCT results was low (≤25%)., Conclusions: NHS publications contain a large number of analyses, the majority of which reported statistically significant but weak associations. Few of these associations have been tested in RCTs, and where they have, the agreement between NHS results and RCTs is poor.

Background and Objectives: Sepsis is a leading cause of mortality and morbidity in the intensive care unit, and many studies have been conducted aiming to improve its outcome. Randomized controlled trials (RCTs) and observational studies using propensity score (PS) method are commonly used for this purpose. However, the agreement between these two major methodological designs has never been investigated in this specific area. The present study aimed to compare the effect sizes between RCTs and PS-based studies., Methods: Electronic databases including Pubmed, Scopus, and EBSCO were searched to obtain PS-based studies in the area of sepsis. The studies were matched to RCTs or systematic reviews and meta-analysis in terms of population, intervention, control, and outcome. When there were multiple PS-based studies or RCTs in one area, the effect sizes were pooled by using random-effects model and inverse variance method. The comparisons were performed by using differences in the effect size., Results: A total of 8 topics were identified fulfilling the criterion that at least 1 pair of RCT and PS-based study could be matched. The interventions included activated protein C, low-dose steroid, antithrombin III, combination antibiotic therapy, fish oil supplementation, statin, etomidate for intubation, and recombinant human soluble thrombomodulin. The effect sizes were statistically different between RCTs and PS-based studies in most circumstances (6/8). The pooled mean difference in effect sizes was -0.16 (95% confidence interval, -0.33 to 0.01), indicating a trend towards larger treatment effect in PS studies than in RCTs. The result remains unaltered by restricting to RCTs and PS studies with the largest sample sizes., Conclusion: Our study shows that PS studies tend to report larger treatment effect than RCTs in the field of sepsis, indicating the difference between efficacy trials and effectiveness studies., (Copyright © 2014 Elsevier Inc. All rights reserved.)

Background: Observation services are provided in greatly variant settings. The aim of this study was to reexamine the effectiveness of observation services compared to inpatient units for patients with nonspecific chest pain., Hypothesis: Patients admitted to observation units have similar outcomes to patients admitted to inpatient wards., Methods: We conducted a claim-based retrospective study for 7549 patients who were admitted to observation and inpatient units. Both models of care were evaluated using the 1-year costs related to chest pain/cardiovascular diseases, and primary and secondary outcomes. Primary outcome was a composite of myocardial infarction, congestive heart failure, stroke, or cardiac arrest, whereas secondary outcomes included revascularization procedures, emergency room revisits, and hospitalization due to cardiovascular diseases., Results: Two-thirds (65.7%, n = 4962) of patients in the sample had observation services, and 34.3% (n = 2587) were admitted to inpatient care. Of the inpatient group, 4.9% experienced a total of 167 primary outcomes, whereas 14.1% experienced a total of 571 secondary outcomes. In comparison, 3.8% of the observation group experienced 238 primary outcomes, and 10.3% experienced 737 secondary outcomes. After adjusting for baseline characteristics using Cox proportional hazard and quantile regression models, no differences between the 2 groups were detected in the 1-year costs of cardiovascular services and primary or secondary outcomes. Patients who had observation services were 79% (95% confidence interval: 1.24-2.58) more likely to have revascularization procedures compared to those admitted to inpatient care., Conclusions: Patients who had observation services had similar outcomes and 1-year costs compared to patients admitted to inpatient wards., (© 2014 Wiley Periodicals, Inc.)

Increasing emphasis has been placed on registries for an organized system used in developing clinical research to improve health care. China has sufficient data that can be applied broadly, but the heterogeneity and irregularity of registries limit their applicability. This article aims to describe the status of registries in China and the related challenges. Patient registries for observational studies were retrieved from the International Clinical Trials Registry to quantitatively evaluate the number of comparatively high-quality registries in China. A literature search was also performed to provide support and updates. A total of 64 patient registries were retrieved from ClinicalTrials.gov using disease, product, and health service as criteria. The sample sizes ranged from 15 to 30,400, with only 12 registries marked as completed. This article describes and compares the detailed information in many aspects. The efficient use of registries has already made considerable progress in China; however, registries still require standardization, high-quality transition, and coordinated development.

Objective: Capturing long-term outcomes from large clinical databases by use of claims data is a potential strategy for improving efficiency while reducing study costs. We sought to compare the use of Medicare data with data from the Women's Health Initiative (WHI) to determine peripheral vascular events, as defined by the WHI study design., Methods: We studied participants from the WHI with both adjudicated outcomes and links to Medicare enrollment and utilization data through 2007. Outcomes of interest included hospitalizations for treatment of abdominal aortic aneurysm (AAA), lower extremity peripheral artery disease (LE PAD), and carotid artery stenosis (CAS). Events determined by WHI adjudication were compared with events defined by coding algorithms using diagnosis and procedure codes from Medicare data with a pilot data set and then validated with a test data set. We assessed agreement by a κ statistic and evaluated reasons for disagreement., Results: In the pilot set, records from 50,511 participants were analyzed. Agreement between the Centers for Medicare and Medicaid Services and WHI for admissions with a diagnosis but no treatment procedures for vascular conditions was poor (κ, 0.02-0.18). On the basis of WHI outcome data collection, vascular treatment procedures occurred in 29 participants for AAA, 204 for LE PAD events, and 281 for CAS. Medicare hospital claims recorded 41 treatments for AAA, 255 for LE PAD, and 317 for CAS. For participants with a Centers for Medicare and Medicaid Services-captured vascular procedure and a record adjudicated by WHI, κ values for treatment procedures were 0.81 for AAA, 0.77 for PAD, and 0.93 for CAS. For vascular procedures identified by WHI but not by Medicare hospital data (n = 82), 55% were captured by Medicare physician claims. Conversely, for treatments identified by Medicare hospital data but not captured by WHI adjudication (n = 57), 74% had physician claims consistent with the procedure. Fifteen participants with AAA or LE PAD procedures in hospital claims had medical records available for review, and nine of these had definitive documentation of procedures that were not captured by the WHI adjudication process. Estimated positive predictive value of Medicare data was 91% to 94% for AAA, 92% to 95% for LE PAD, and 94% to 99% for CAS. Available test set data (n = 50,253) yielded generally similar results with κ of 0.77 for AAA, 0.79 for LE PAD, and 0.94 for CAS., Conclusions: Medicare data appear useful for identifying vascular treatment procedures for WHI participants. Medicare hospital claims identify more procedures than WHI does, with high positive predictive value, but also may not capture some procedures identified in WHI., (Copyright © 2014. Published by Mosby, Inc.)

Background: Observational studies dominate the surgical literature. Statistical adjustment is an important strategy to account for confounders in observational studies. Research has shown that published articles are often poor in statistical quality, which may jeopardize their conclusions. The Statistical Analyses and Methods in the Published Literature (SAMPL) guidelines have been published to help establish standards for statistical reporting.This study will seek to determine whether the quality of statistical adjustment and the reporting of these methods are adequate in surgical observational studies. We hypothesize that incomplete reporting will be found in all surgical observational studies, and that the quality and reporting of these methods will be of lower quality in surgical journals when compared with medical journals. Finally, this work will seek to identify predictors of high-quality reporting., Methods/design: This work will examine the top five general surgical and medical journals, based on a 5-year impact factor (2007-2012). All observational studies investigating an intervention related to an essential component area of general surgery (defined by the American Board of Surgery), with an exposure, outcome, and comparator, will be included in this systematic review. Essential elements related to statistical reporting and quality were extracted from the SAMPL guidelines and include domains such as intent of analysis, primary analysis, multiple comparisons, numbers and descriptive statistics, association and correlation analyses, linear regression, logistic regression, Cox proportional hazard analysis, analysis of variance, survival analysis, propensity analysis, and independent and correlated analyses. Each article will be scored as a proportion based on fulfilling criteria in relevant analyses used in the study. A logistic regression model will be built to identify variables associated with high-quality reporting. A comparison will be made between the scores of surgical observational studies published in medical versus surgical journals. Secondary outcomes will pertain to individual domains of analysis. Sensitivity analyses will be conducted., Discussion: This study will explore the reporting and quality of statistical analyses in surgical observational studies published in the most referenced surgical and medical journals in 2013 and examine whether variables (including the type of journal) can predict high-quality reporting.

Background: Results from cohort studies evaluating the severity of respiratory viral co-infections are conflicting. We conducted a systematic review and meta-analysis to assess the clinical severity of viral co-infections as compared to single viral respiratory infections., Methods: We searched electronic databases and other sources for studies published up to January 28, 2013. We included observational studies on inpatients with respiratory illnesses comparing the clinical severity of viral co-infections to single viral infections as detected by molecular assays. The primary outcome reflecting clinical disease severity was length of hospital stay (LOS). A random-effects model was used to conduct the meta-analyses., Results: Twenty-one studies involving 4,280 patients were included. The overall quality of evidence applying the GRADE approach ranged from moderate for oxygen requirements to low for all other outcomes. No significant differences in length of hospital stay (LOS) (mean difference (MD) -0.20 days, 95% CI -0.94, 0.53, p = 0.59), or mortality (RR 2.44, 95% CI 0.86, 6.91, p = 0.09) were documented in subjects with viral co-infections compared to those with a single viral infection. There was no evidence for differences in effects across age subgroups in post hoc analyses with the exception of the higher mortality in preschool children (RR 9.82, 95% CI 3.09, 31.20, p<0.001) with viral co-infection as compared to other age groups (I2 for subgroup analysis 64%, p = 0.04)., Conclusions: No differences in clinical disease severity between viral co-infections and single respiratory infections were documented. The suggested increased risk of mortality observed amongst children with viral co-infections requires further investigation.

A goal of many health studies is to determine the causal effect of a treatment or intervention on health outcomes. Often, it is not ethically or practically possible to conduct a perfectly randomized experiment, and instead, an observational study must be used. A major challenge to the validity of observational studies is the possibility of unmeasured confounding (i.e., unmeasured ways in which the treatment and control groups differ before treatment administration, which also affect the outcome). Instrumental variables analysis is a method for controlling for unmeasured confounding. This type of analysis requires the measurement of a valid instrumental variable, which is a variable that (i) is independent of the unmeasured confounding; (ii) affects the treatment; and (iii) affects the outcome only indirectly through its effect on the treatment. This tutorial discusses the types of causal effects that can be estimated by instrumental variables analysis; the assumptions needed for instrumental variables analysis to provide valid estimates of causal effects and sensitivity analysis for those assumptions; methods of estimation of causal effects using instrumental variables; and sources of instrumental variables in health studies., (Copyright © 2014 John Wiley & Sons, Ltd.)

Background: The aim of this study was to examine the clinical characteristics of rheumatoid arthritis (RA) patients who underwent cervical spine surgery using a multicenter observational database., Methods: We obtained data from a nationwide observational cohort database of patients with rheumatic diseases (National Database of Rheumatic Diseases by iR-net in Japan (NinJa)) for the fiscal years 2003 to 2011. A total of 39 out of 60 patients who underwent cervical spine surgery for a RA-related cause and whose data were available for two consecutive years (to assess the preoperative patient status) were chosen as cases. Patients with a non-RA-related cause of surgery (e.g., trauma) were excluded. First, we compared the patient characteristics between the cases and total patients in the same fiscal year. Next, 106 eligible controls, who were defined as RA patients enrolled in the same fiscal year as the case subjects, who were matched for age, gender and disease duration (within ±1 year), were selected. We compared the demographic data between the two groups. We also calculated the percentage of patients who underwent cervical spine surgery (surgeries/total number of patients) in fiscal years 2003 to 2011., Results: Although the proportion of patients using biologics linearly increased during study period, the percentage of patients undergoing cervical spine surgeries remained unchanged, at approximately 0.15%. These cases had more tender joints (3 vs. 1, p < 0.01) and exhibited a significantly higher Modified Health Assessment Questionnaire (MHAQ) score (1.13 vs. 0.5, p < 0.01), C-reactive protein (CRP) (1.5 vs. 0.36, p < 0.01), and disease activity score (DAS) 28-CRP (3.63 vs. 2.81, p < 0.01) compared to the controls., Conclusions: Our study revealed that RA patients requiring cervical spine surgery have a higher disease activity (as represented by the DAS28-CRP) and are more functionally disabled (as represented by the MHAQ) than control patients.

Missing covariate data often arise in biomedical studies, and analysis of such data that ignores subjects with incomplete information may lead to inefficient and possibly biased estimates. A great deal of attention has been paid to handling a single missing covariate or a monotone pattern of missing data when the missingness mechanism is missing at random. In this article, we propose a semiparametric method for handling non-monotone patterns of missing data. The proposed method relies on the assumption that the missingness mechanism of a variable does not depend on the missing variable itself but may depend on the other missing variables. This mechanism is somewhat less general than the completely non-ignorable mechanism but is sometimes more flexible than the missing at random mechanism where the missingness mechansim is allowed to depend only on the completely observed variables. The proposed approach is robust to misspecification of the distribution of the missing covariates, and the proposed mechanism helps to nullify (or reduce) the problems due to non-identifiability that result from the non-ignorable missingness mechanism. The asymptotic properties of the proposed estimator are derived. Finite sample performance is assessed through simulation studies. Finally, for the purpose of illustration we analyze an endometrial cancer dataset and a hip fracture dataset., (© 2014, The International Biometric Society.)

Suppose we are interested in estimating the average causal effect from an observational study. A doubly robust estimator, which is a hybrid of the outcome regression and propensity score weighting, is more robust than estimators obtained by either of them in the sense that, if at least one of the two models holds, the doubly robust estimator is consistent. However, a doubly robust estimator may still suffer from model misspecification since it is not consistent if neither of them is correctly specified. In this article, we propose an alternative estimator, called the stratified doubly robust estimator, by further combining propensity score stratification with outcome regression and propensity score weighting. This estimator allows two candidate models for the propensity score and is more robust than existing doubly robust estimators in the sense that it is consistent either if the outcome regression holds or if one of the two models for the propensity score holds. Asymptotic properties are examined and finite sample performance of the proposed estimator is investigated by simulation studies. Our proposed method is illustrated with the Tone study, which is a community survey conducted in Japan., (© 2014, The International Biometric Society.)