Assessment of covariate effects in Aalen's additive hazard model

We study the properties of test statistics for a covariate effect in Aalen's additive hazard model and propose several new test statistics. The proposed statistics are derived by using the weights from linear rank statistics for comparing two survival curves. We compare these statistics with the two statistics proposed by Aalen using Monte Carlo simulations. Several different survival configurations are considered in the simulation study: proportional hazards; crossing hazards; hazard differences early in time, and hazard differences for large survival times. Of the proposed test statistics, one is superior for detecting hazard differences for large survival times and another is superior for detecting early hazard differences and crossing hazards. © 1998 John Wiley & Sons, Ltd.     

Moral hazard and selection for voluntary deductibles

This paper investigates whether the voluntary deductible in the Dutch health insurance system reduces moral hazard or acts only as a cost reduction tool for low‐risk individuals. We use a sample of 14,089 observations, comprising 2,939 individuals over seven waves from the Longitudinal Internet Studies for the Social sciences panel for the analysis. We employ bivariate models that jointly model the choice of a deductible and health care utilization and supplement the identification with an instrumental variable strategy. The results show that the voluntary deductible reduces moral hazard, especially in the decision to visit a doctor (extensive margin) compared with the number of visits (intensive margin). In addition, a robustness test shows that selection on moral hazard is not present in this context.     

Sample size determination for jointly testing a cause‐specific hazard and the all‐cause hazard in the presence of competing risks

This article considers sample size determination for jointly testing a cause‐specific hazard and the all‐cause hazard for competing risks data. The cause‐specific hazard and the all‐cause hazard jointly characterize important study end points such as the disease‐specific survival and overall survival, which are commonly used as coprimary end points in clinical trials. Specifically, we derive sample size calculation methods for 2‐group comparisons based on an asymptotic chi‐square joint test and a maximum joint test of the aforementioned quantities, taking into account censoring due to lost to follow‐up as well as staggered entry and administrative censoring. We illustrate the application of the proposed methods using the Die Deutsche Diabetes Dialyse Studies clinical trial. An R package “powerCompRisk” has been developed and made available at the CRAN R library.     

Estimating adverse selection and moral hazard effects with hospital invoices data in a government-controlled healthcare system

We use invoices for hospital services from a regional hospital in Croatia to test for adverse selection and moral hazard. There are three categories of patients: with no supplemental insurance, who bought it, and who are entitled to it for free. Our identification procedure relies on the premise that the difference in the observed medical care consumption between the patients who bought the insurance and those entitled to free insurance is caused by pure selection effect, whereas the difference in healthcare consumption between the group that received the free insurance and the group that has no insurance is due to moral hazard. Results show favorable selection for patients in 20- to 30-year-old cohort and significant moral hazard for all age cohorts. The selection effect reverses its sign in older cohorts explained by the differences in risk aversion across cohorts caused by the timing of transition from socialism to market economy.     

Tutorial on discrete hazard functions

Risk analysis requires estimation of hazard functions. A hazard rate is the conditional probability of adverse sentinel event occurring in the next time period, given that it has not yet occurred. This tutorial shows how hazard functions are estimated from survival functions, the probability of going through a time period without the sentinel event. Survival functions are built on cumulative distribution functions, which measure the probability of occurrence of sentinel event in current and prior time periods. Cumulative distribution functions are calculated from probability density functions, which give the probability of an event occurring at a particular time period. Probability density functions are typically estimated from incidence reports, which are readily available to safety officers. Sometimes, these functions are estimated by making assumptions about the shape of the distribution function. For discrete data, the typical probability density functions are Bernoulli, Binominal, Geometric, and Poisson distributions. This tutorial starts with estimating a probability distribution and then proceeds to calculation of hazard and relative risk rates.     

Estimating HIV hazard rates from cross-sectional HIV prevalence data

We propose a method of estimating the HIV hazard rate in a population when only cross-sectional data measuring HIV-prevalence are available. In a cross-sectional study, individuals who test positive are left-censored since we know only that the transition time is smaller than the monitoring time. Individuals who test negative are right-censored since we know only that the transition time is greater than the monitoring time. The transition time is not observed directly. For our method, we assume the hazard rate is well-defined after the time we start screening. We assume further the mortality rates for HIV-positive and HIV-negative individuals are known. The method is illustrated using HIV-prevalence data collected over four years among women in Cape Town, South Africa.     

A balanced hazard ratio for risk group evaluation from survival data

Common clinical studies assess the quality of prognostic factors, such as gene expression signatures, clinical variables or environmental factors, and cluster patients into various risk groups. Typical examples include cancer clinical trials where patients are clustered into high or low risk groups. Whenever applied to survival data analysis, such groups are intended to represent patients with similar survival odds and to select the most appropriate therapy accordingly. The relevance of such risk groups, and of the related prognostic factors, is typically assessed through the computation of a hazard ratio. We first stress three limitations of assessing risk groups through the hazard ratio: (1) it may promote the definition of arbitrarily unbalanced risk groups; (2) an apparently optimal group hazard ratio can be largely inconsistent with the p‐value commonly associated to it; and (3) some marginal changes between risk group proportions may lead to highly different hazard ratio values. Those issues could lead to inappropriate comparisons between various prognostic factors. Next, we propose the balanced hazard ratio to solve those issues. This new performance metric keeps an intuitive interpretation and is as simple to compute. We also show how the balanced hazard ratio leads to a natural cut‐off choice to define risk groups from continuous risk scores. The proposed methodology is validated through controlled experiments for which a prescribed cut‐off value is defined by design. Further results are also reported on several cancer prognosis studies, and the proposed methodology could be applied more generally to assess the quality of any prognostic markers. Copyright © 2015 John Wiley & Sons, Ltd.     

信息不对称条件下医生道德风险的防范

文章在对医疗服务市场进行经济学分析的基础上，阐述了医生道德风险的表现形式和存在原因，并从法规建设、医德教育、激励机制、人性化管理、医患关系和医院文化等方面探讨了防范道德风险的措施。     

危险源风险体积法分析探索

[目的]以"风险体积"概括和展现危险源风险构成的整体性,探索破解危险源风险的理论途径。[方法]从经典风险理论,导出"风险体积"的概念,对危险源存在的事故风险进行三维分析,界定出危险源风险与隐患,体现从危险源到隐患再到风险的溯源关系。[结果]提出"一元隐患"、"二元隐患"和"三元风险"的观点。[结论]对于风险体积(三元风险),可通过降低(或消除)其中一种构成因素,使其缩小(或为零),事故风险下降(或消失)。     

A comparison of two simple hazard ratio estimators based on the logrank test

Two hazard ratio estimators based on the logrank test are investigated using a simulation study. The Pike estimator (ratio of relative death rates) was shown to be consistently less biased than the Peto (1-step) estimator. The latter has recently been advocated as the method of choice for point and interval estimation. Both estimators exhibited bias with increasing hazard ratios, although the bias was minimal for effects less than 3. The confidence intervals also did not achieve the nominal coverage with increasing hazard ratios, but again the Pike estimator was superior. The coverage could be improved by recalculation of the variance incorporating the point estimate. For a hazard ratio of less than 3 we recommend the use of the Pike estimator, otherwise it is necessary to use a more complex method of estimation.     

An overview of process hazard evaluation techniques

Since the 1985 release of methyl isocyanate in Bhopal, India, which killed thousands, the chemical industry has begun to use process hazard analysis techniques more widely to protect the public from catastrophic chemical releases. These techniques can provide a systematic method for evaluating a system design to ensure that it operates as intended, help identify process areas that may result in the release of a hazardous chemical, and help suggest modifications to improve process safety. Eight different techniques are discussed, with some simple examples of how they might be applied. These techniques include checklists, "what if" analysis, safety audits and reviews, preliminary hazard analysis (PHA), failure modes and effect analysis (FMEA), fault tree analysis (FTA), event tree analysis (ETA), and hazard and operability studies (HAZOP). The techniques vary in sophistication and scope, and no single one will always be the best. These techniques can also provide the industrial hygienist with the tools needed to protect both workers and the community from both major and small-scale chemical releases. A typical industrial hygiene evaluation of a facility would normally include air sampling. If the air sampling does detect a specific hazardous substance, the source will probably be a routine or continuous emission. However, air sampling will not be able to identify or predict the location of a nonroutine emission reliably. By incorporating these techniques with typical evaluations, however, industrial hygienists can proactively help reduce the hazards to the workers they serve.     

Proportional hazard model estimation under dependent censoring using copulas and penalized likelihood

This paper considers Cox proportional hazard models estimation under informative right censored data using maximum penalized likelihood, where dependence between censoring and event times are modelled by a copula function and a roughness penalty function is used to restrain the baseline hazard as a smooth function. Since the baseline hazard is nonnegative, we propose a special algorithm where each iteration involves updating regression coefficients by the Newton algorithm and baseline hazard by the multiplicative iterative algorithm. The asymptotic properties for both regression coefficients and baseline hazard estimates are developed. The simulation study investigates the performance of our method and also compares it with an existing maximum likelihood method. We apply the proposed method to a dementia patients dataset.     

Estimation of a decreasing hazard of patients with acute coronary syndrome

The Kaplan–Meier, Nelson–Aalen and Breslow estimators are widely used in the analysis of right‐censored time to event data in medical applications. These methods are fully non‐parametric and do not put any restriction on the shape of the hazard curve. In some applications, this leads to implausible estimates of the hazard course over time. With non‐parametric shape‐constrained estimation techniques, one can facilitate an increasing or decreasing hazard and thus generate estimators that better match the biological reasoning, without being as restrictive as parametric methods. We illustrate the advantage of such techniques in the analysis of a large clinical trial in cardiology. Simulation results show that in case the true hazard is monotone, the non‐parametric shape‐constrained estimators are more accurate than the traditional estimators on the hazard level. On the (cumulative) distribution function level, the shape‐constrained estimators show similar performance as the traditional ones. Copyright © 2012 John Wiley & Sons, Ltd.     

Health and hazard surveillance--needs and perspectives

Health and hazard surveillance is the on-going collection, analysis, evaluation, and dissemination of information aimed at improving the health, work ability, and well-being of workers. Hazard surveillance supplemented by quantitative risk assessment is the most prevention-oriented domain of surveillance. However, information on exposures and stress factors is often lacking. Computer-assisted expert judgment is a new approach to improve knowledge on hazards. Surveillance of work organizations and psychosocial factors is a challenging area which still requires research. Statistics on occupational injuries and diseases are available, but nonspecific coding and insufficient interpretation often limit their usefulness. Systematic surveys of new types of cases and the linking of different data bases are becoming more feasible. Methodological work is needed to develop indicators of work ability and work-related symptoms. Among the many possibilities to develop surveillance are the use of information technology, the application of hazard communication principles, and the development of practical aids for surveillance at the company level.     

Rapid experimental rationale for a waste hazard class by cytotoxicity

The paper presents the results of experimental and analytical studies substantiating a classification of waste hazard by the cytotoxicity indices. The authors have established a significant correlation between the substance toxicity values obtained in vivo and in vitro and show it possible to make an approximate forecast of the average lethal concentration of substances by the estimates made on cell cultures. The criteria for toxicological waste hazard, which are adequate to those for the hazards of chemicals by DL50, are given.     

医院道德风险管理与防范

医疗行业是一个特殊的行业，要求从业人员必须具备较高的道德水准，必须以救死扶伤、治病救人为最高行为准则，否则就会引发道德风险乃至危机。因此，医院的道德风险管理及防范已经成为必须引起重视和解决的一个极为重要的医院管理问题。本文从医院道德风险的概念，道德风险的表现形式，道德风险产生的原因分析，道德风险的管理与防范四个方面进行了阐述。     

Drug price regulation under consumer moral hazard

Drug price differences across national markets as they exist in the EU are often justified by the concept of Ramsey prices: with fixed costs for R&D, the optimal mark-ups on marginal costs are inversely related to the price elasticity in the individual markets. This well-known result prevails if consumer moral hazard is taken into account. Contrary to the situation without moral hazard, the uniform price does not necessarily dominate discriminatory pricing in welfare terms. The two-part tariff is a better alternative as it allows governments to address moral hazard. A uniform price combined with lump-payments reflecting differences in the willingness to pay and the moral hazard in member states appears to be an attractive option for a common EU drug market.     

The effect of lead-based paint hazard remediation on blood lead levels of lead poisoned children in New York City

Despite the widespread use of lead paint hazard control for children with lead poisoning, few controlled studies that estimate the effect of such control on children's blood lead levels have been published. This retrospective follow-up study examined the effects of lead hazard remediation and its timing on the blood lead levels of lead-poisoned children. From the New York City child blood lead registry, 221 children were selected who had an initial blood lead level of 20-44 micro g/dL between 1 July 1994 and 31 December 1996; were 6 months to 6 years of age; had a report of a follow-up blood lead test between 10 and 14 months after the initial test; had a lead-based paint hazard identified in the primary dwelling unit prior to the 10- to 14-month follow-up blood lead test; had resided or spent time at only one address with an identified lead-based paint hazard; and were not chelated. The decline in geometric mean blood lead levels from baseline to 10-14 months later was compared for children whose homes were remediated and whose homes were not remediated during the follow-up period. Regardless of remediation, geometric mean blood lead levels declined significantly from 24.3 micro g/dL at the initial diagnosis to 12.3 micro g/dL at the 10- to 14-month follow-up blood lead test (P<0.01). Among the 146 children whose homes were remediated the geometric mean blood lead levels declined 53% compared to 41% among the 75 children whose homes were not remediated by the follow-up blood lead test, a remediation effect of approximately 20% (P<0.01). After adjusting for potential confounders, the remediation effect was 11%, although it was no longer significant. Race was the only factor that appeared to confound the relationship: Black children had higher follow-up blood lead levels even after controlling for other factors, including the natural logarithm of the initial blood lead level. The effect of remediation appeared to be stronger for younger (10 to <36 months old) than for older (36 to 72 months old) children (P=0.06). While children in homes with earlier remediation (within less than 3 months) appeared to have greater declines in blood lead levels at the follow-up test than children in homes with later remediation (after 3 or more months), this trend was not significant when controlling for confounding factors. The findings of this study suggest that early identification of lead-poisoned children and timely investigation and abatement of hazards contribute to reducing blood lead levels. However, the apparent effect is modest and further research is needed to systematically test and improve the effectiveness of lead hazard controls.     

Field laser hazard calculations

The authors develop formulae for laser hazard evaluation. Beam diameter, transmitted power through an optical system, nominal ocular hazard distance, and optical density required for eye protection may be calculated. A calculator program is provided which uses experimentally determined laser parameters to return the necessary safety information.