Constrained empirical-likelihood confidence regions in nonignorable covariate-missing data problems

Missing covariates in regression analysis are a pervasive problem in medical, social, and economic researches. We study empirical-likelihood confidence regions for unconstrained and constrained regression parameters in a nonignorable covariate-missing data problem. For an assumed conditional mean regression model, we assume that some covariates are fully observed but other covariates are missing for some subjects. By exploitation of a probability model of missingness and a working conditional score model from a semiparametric perspective, we build a system of unbiased estimating equations, where the number of equations exceeds the number of unknown parameters. Based on the proposed estimating equations, we introduce unconstrained and constrained empirical-likelihood ratio statistics to construct empirical-likelihood confidence regions for the underlying regression parameters without and with constraints. We establish the asymptotic distributions of the proposed empirical-likelihood ratio statistics. Simulation results show that the proposed empirical-likelihood methods have a better finite-sample performance than other competitors in terms of coverage probability and interval length. Finally, we apply the proposed empirical-likelihood methods to the analysis of a data set from the US National Health and Nutrition Examination Survey.     

Long-term effectiveness of zoster vaccine live for postherpetic neuralgia prevention

BackgroundPostherpetic neuralgia (PHN) occurs in 5–30% of individuals with herpes zoster (HZ) and is characterized by long-lasting pain. Zoster vaccine live (ZVL) is licensed for people 50 years and older to prevent HZ and PHN. This study evaluated vaccine effectiveness (VE) of ZVL against PHN.MethodsWe conducted an open cohort study within Kaiser Permanente Northern California with continuous accrual of people as they became age-eligible for ZVL. We defined PHN using a PHN diagnosis between 90 and 365 days after an incident episode of HZ. We estimated VE against PHN using Cox regression with a calendar timeline stratified by year of birth and adjusted for sex, race, influenza vaccination, outpatient visit frequency, comorbidities, and immune compromise status.ResultsFrom 2007 to 2016, 1·5 million people entered the study population and 33% received ZVL. During 7·6 million person-years of follow-up, there were 62,205 HZ cases, 4150 (6·7%) of which went on to develop PHN. Overall VE for PHN was 64·8% (95% CI 61·3, 68). VE was 82·8% (95% CI 77·6, 86·7) during the first year after vaccination, 58·3% (95% CI 50.1, 65.2) during the third year, and then waned more gradually to 48·7% (95% CI 30·2, 62·3) during the eighth year. VE in persons vaccinated when aged 80 years or older was similar to VE in younger vaccinees. VE in persons vaccinated when immune compromised was similar to VE in immune competent.ConclusionsOverall, ZVL was 65% effective against PHN. It was effective in all age groups and provided moderate protection through 8 years.     

A case-control study of risk factors for intussusception among infants in eastern France after the introduction of the rotavirus vaccine

ObjectiveThe objective of the present study was to investigate the risk factors for intussusception (IS) among infants, including vaccination against rotavirus.MethodsCase-control study with systematic inclusion of all infants aged <1 year with suspected IS admitted to emergency departments in the eastern region of France between 1 April 2008 and 31 March 2012. All cases classed level 1 according to the Brighton classification were matched to 4 hospital controls. Two exposure windows were examined; exposure to the first dose of rotavirus vaccine in the 7 and in the 14 days prior to the occurrence of IS.ResultsA total of 115 cases were matched with 457 controls. The average vaccination coverage rate over the 4 years of study was 8.6%. Rotavirus vaccine was not found to be significantly associated with the occurrence of IS in the 7 days (odds ratio (OR) not calculated; p = 0.99) and in the 14 days after administration of one dose vaccine (OR 1.33, 95% confidence interval (CI) 0.14–12.82). Infant formula alone or combined with breastfeeding was associated with an excess risk of IS (OR 2.74, 95% CI 1.10–6.79). A history of gastroenteritis within 2 weeks prior to hospitalisation was also associated with an increased risk (OR 2.24, 95% CI 1.07–4.67).ConclusionOur study indicates that infant formula alone or combined with breastfeeding is a risk factor for IS. A small, non-significant increase in the risk of IS was observed after rotavirus vaccination, although the low vaccine coverage rate likely precluded detection of a significant increase in risk.     

The formation of physician patient sharing networks in medicare: Exploring the effect of hospital affiliation

This study explores the forces that drive the formation of physician patient sharing networks. In particular, I examine the degree to which hospital affiliation drives physicians' sharing of Medicare patients. Using a revealed preference framework where observed network links are taken to be pairwise stable, I estimate the physicians' pair‐specific values using a tetrad maximum score estimator that is robust to the presence of unobserved physician specific characteristics. I also control for a number of potentially confounding patient sharing channels, such as (a) common physician group or hospital system affiliation, (b) physician homophily, (c) knowledge complementarity, (d) patient side considerations related to both geographic proximity and insurance network participation, and (e) spillover from other collaborations. Focusing on the Chicago hospital referral region, I find that shared hospital affiliation accounts for 36.5% of the average pair‐specific utility from a link. Implications for reducing care fragmentation are discussed.     

Chronic inflammation and risk of lung cancer in older adults in the health,aging and body composition cohort study

Objectives

We examined the association between three inflammatory markers (Interleukin (IL)-6, C-reactive protein (CRP), tumor necrosis factor (TNF)-α) and incident lung cancer using baseline, updated, and averaged inflammatory measures in older adults.

The efficacy of DNA mixture to mixture matching

Standard practice in forensic science is to compare a person of interest’s (POI) reference DNA profile with an evidence DNA profile and calculate a likelihood ratio that considers propositions including and excluding the POI as a DNA donor. A method has recently been published that provides the ability to compare two evidence profiles (of any number of contributors and of any level of resolution) comparing propositions that consider the profiles either have a common contributor, or do not have any common contributors. Using this method, forensic analysts can provide intelligence to law enforcement by linking crime scenes when no suspects may be available. The method could also be used as a quality assurance measure to identify potential sample to sample contamination. In this work we analyse a number of constructed mixtures, ranging from two to five contributors, and with known numbers of common contributors, in order to investigate the performance of using likelihood ratios for mixture to mixture comparisons. Our findings demonstrate the ability to identify common donors in DNA mixtures with the power of discrimination depending largely on the least informative mixture of the pair being considered. The ability to match mixtures to mixtures may provide intelligence information to investigators by identifying possible links between cases which otherwise may not have been considered connected.