Effects of informative censoring on the hazard function: Application to the proportional hazards regression model

The standard analyses of survival data involve the assumption that survival and censoring are independent. When censoring and survival are related, the phenomenon is known as informative censoring. This paper examines the effects of an informative censoring assumption on the hazard function and the estimated hazard ratio provided by the Cox model.^ The limiting factor in all analyses of informative censoring is the problem of non-identifiability. Non-identifiability implies that it is impossible to distinguish a situation in which censoring and death are independent from one in which there is dependence. However, it is possible that informative censoring occurs. Examination of the literature indicates how others have approached the problem and covers the relevant theoretical background.^ Three models are examined in detail. The first model uses conditionally independent marginal hazards to obtain the unconditional survival function and hazards. The second model is based on the Gumbel Type A method for combining independent marginal distributions into bivariate distributions using a dependency parameter. Finally, a formulation based on a compartmental model is presented and its results described. For the latter two approaches, the resulting hazard is used in the Cox model in a simulation study.^ The unconditional survival distribution formed from the first model involves dependency, but the crude hazard resulting from this unconditional distribution is identical to the marginal hazard, and inferences based on the hazard are valid. The hazard ratios formed from two distributions following the Gumbel Type A model are biased by a factor dependent on the amount of censoring in the two populations and the strength of the dependency of death and censoring in the two populations. The Cox model estimates this biased hazard ratio. In general, the hazard resulting from the compartmental model is not constant, even if the individual marginal hazards are constant, unless censoring is non-informative. The hazard ratio tends to a specific limit.^ Methods of evaluating situations in which informative censoring is present are described, and the relative utility of the three models examined is discussed. ^

Assessing time-by-covariate interactions in Cox proportional hazards regression models using cubic spline functions

This dissertation develops and explores the methodology for the use of cubic spline functions in assessing time-by-covariate interactions in Cox proportional hazards regression models. These interactions indicate violations of the proportional hazards assumption of the Cox model. Use of cubic spline functions allows for the investigation of the shape of a possible covariate time-dependence without having to specify a particular functional form. Cubic spline functions yield both a graphical method and a formal test for the proportional hazards assumption as well as a test of the nonlinearity of the time-by-covariate interaction. Five existing methods for assessing violations of the proportional hazards assumption are reviewed and applied along with cubic splines to three well known two-sample datasets. An additional dataset with three covariates is used to explore the use of cubic spline functions in a more general setting. ^

A SIMULATION STUDY OF THE SIZE AND POWER OF SOME TWO-SAMPLE TESTS FOR UNEQUALLY CENSORED SURVIVAL DATA (RANK TESTS, LOGRANK, PROPORTIONAL HAZARDS, WILCOXON TESTS, EXPONENTIAL DISTRIBUTION)

Sizes and power of selected two-sample tests of the equality of survival distributions are compared by simulation for small samples from unequally, randomly-censored exponential distributions. The tests investigated include parametric tests (F, Score, Likelihood, Asymptotic), logrank tests (Mantel, Peto-Peto), and Wilcoxon-Type tests (Gehan, Prentice). Equal sized samples, n = 18, 16, 32 with 1000 (size) and 500 (power) simulation trials, are compared for 16 combinations of the censoring proportions 0%, 20%, 40%, and 60%. For n = 8 and 16, the Asymptotic, Peto-Peto, and Wilcoxon tests perform at nominal 5% size expectations, but the F, Score and Mantel tests exceeded 5% size confidence limits for 1/3 of the censoring combinations. For n = 32, all tests showed proper size, with the Peto-Peto test most conservative in the presence of unequal censoring. Powers of all tests are compared for exponential hazard ratios of 1.4 and 2.0. There is little difference in power characteristics of the tests within the classes of tests considered. The Mantel test showed 90% to 95% power efficiency relative to parametric tests. Wilcoxon-type tests have the lowest relative power but are robust to differential censoring patterns. A modified Peto-Peto test shows power comparable to the Mantel test. For n = 32, a specific Weibull-exponential comparison of crossing survival curves suggests that the relative powers of logrank and Wilcoxon-type tests are dependent on the scale parameter of the Weibull distribution. Wilcoxon-type tests appear more powerful than logrank tests in the case of late-crossing and less powerful for early-crossing survival curves. Guidelines for the appropriate selection of two-sample tests are given. ^

La cultura política en Corrientes

En esta tesis se aborda la historia política correntina del período 1909- 1930, centrándose en la actuación de los partidos políticos, en las elecciones y prácticas políticas y electorales con el objeto de contribuir a la reconstrucción y comprensión de la cultura política provincial y aportar al conocimiento de la historia política argentina de esa etapa. La provincia de Corrientes, constituye un caso peculiar en el contexto nacional, al convertirse en la única que no tuvo un gobierno radical en todo el período posterior a la sanción de la ley electoral de 1912. Lo que aquí se trató de demostrar fue que los mecanismos que permitieron la permanencia de los gobiernos conservadores en esa provincia fueron tanto institucionales como culturales. Entre ellos, se destacan la adecuación al nuevo sistema político por medio de una reforma de la Constitución Provincial que confirmó el sistema de representación proporcional en lugar del sistema de lista incompleta y la reorganización de los partidos conservadores a través de la instauración de la política del acuerdo. Así, la permanencia de éste sector en el gobierno durante todo el período analizado, estuvo más vinculada con la puesta en práctica de estos mecanismos, que con el incremento de la participación del electorado o la influencia de elementos discursivos o ideológicos.

It's confusing when the Swiss go to the polls. Here are 6 surprising things about their elections

On Sunday, Switzerland votes on its federal parliament. The members of both chambers - 200 National Councillors and 46 Councillors of States - are directly elected, with the 26 Swiss cantons (or provinces) forming the districts. Seats in the National Council are distributed to the cantons based on their population size, while in the Council of States each full canton gets two seats and the six half-cantons get one each. Here are six things to look out for.

The 2015 Swiss elections: A landslide win for the right, despite limited changes in vote share

Switzerland held federal elections on 18 October, with the conservative Swiss People’s Party winning the largest share of the votes. Daniel Bochsler, Marlène Gerber and David Zumbach write that while the increase in vote share for the Swiss People’s Party was relatively limited, the party managed to significantly increase the number of seats it holds in Switzerland’s lower house of parliament, the National Council. Nevertheless, the party is unlikely to make substantial gains in the country’s upper house, the Senate, as it traditionally struggles under the two-round electoral system used in Senate elections.

Political conditions for fair elections

Democracy is not necessarily consolidated simply by the introduction of formal democratic institutions. It is often observed in new democracies that democratic institutions are neglected and eroded in actual practice. Particularly, electoral fraud committed by a ruler is one of the main problems in this regard. This paper deals with two questions, (1) under what conditions does a ruler have an incentive to hold fair elections (or to rig elections), and (2) what makes a ruler prefer to establish an independent election governing institution? Assuming that a ruler prefers to maintain her power, basically she has an incentive to rig elections in order to be victorious in the political competition. A ruler, however, faces the risk of losing power if the opposition stages successful protests on a sufficiently large scale. If opponents are able to pose a credible threat to a ruler, she will have an incentive to hold fair elections. The problem is that information on electoral fraud is not shared by every player in the game. For the opposition, imperfect information deepens their coordination problems. Imperfect information, on the other hand, in some cases causes a problem for a ruler. If the opposition is sufficiently cohesive and have little tolerance of cheating, even unverified suspicions of fraud may trigger menacing protests. In such a case, a ruler has an incentive to establish an independent election commission to avoid unnecessary collisions by revealing the nature of the elections.

Ethnicity and Elections under Authoritarianism: The Case of Kazakhstan

Despite the ethnicisation of power since independence in 1991, Kazakhstan has managed to maintain political stability without experiencing large-scale mobilisation to oppose Kazakh domination. This paper examines government strategy to avoid ethnic voting in an attempt to explain why ethnic divisions were rarely reflected in the struggle for power in the republic. While the arbitrary use of legal provisions considerably limited participation in elections by ethnic leaders, powerful pro-president parties that exhibited a cross-ethnic character were created to curtail ethnically based movements. The control strategy in elections aimed not simply at ethnicising the parliament in favour of Kazakhs, but at having loyal Russians and other minorities represented in the legislature through nomination by the president and catch-all pro-regime parties, or through the presidential consultative body—Assembly of the People of Kazakhstan. This well-controlled representation of minorities served not only to placate non-Kazakhs but also to provide legitimacy for the Kazakh-dominated leadership by projecting the image of cross-ethnic support for the president and some degree of power-sharing.

Changing local elite selection in Thailand : emergence of new local government presidents after direct elections and their capabilities

Before rural local government units were established in Thailand, reform debates within the country faced a crucial issue: Candidates at the rural sub-district levels might adopt electioneering methods such as vote buying and the patronage system of the local political and economic elite, the methods that had been used in the national elections. In fact, the results of the 2006 survey in this paper, which followed the introduction of direct elections in rural local government units in 2003, contrast with the result anticipated during the debates on political reform. The preliminary data of the survey shows that the decentralization process and the introduction of the direct election system in the rural areas had some effect in changing the selection process of the local elite in Thailand.

Genetic instability of cancer cells is proportional to their degree of aneuploidy

Genetic and phenotypic instability are hallmarks of cancer cells, but their cause is not clear. The leading hypothesis suggests that a poorly defined gene mutation generates genetic instability and that some of many subsequent mutations then cause cancer. Here we investigate the hypothesis that genetic instability of cancer cells is caused by aneuploidy, an abnormal balance of chromosomes. Because symmetrical segregation of chromosomes depends on exactly two copies of mitosis genes, aneuploidy involving chromosomes with mitosis genes will destabilize the karyotype. The hypothesis predicts that the degree of genetic instability should be proportional to the degree of aneuploidy. Thus it should be difficult, if not impossible, to maintain the particular karyotype of a highly aneuploid cancer cell on clonal propagation. This prediction was confirmed with clonal cultures of chemically transformed, aneuploid Chinese hamster embryo cells. It was found that the higher the ploidy factor of a clone, the more unstable was its karyotype. The ploidy factor is the quotient of the modal chromosome number divided by the normal number of the species. Transformed Chinese hamster embryo cells with a ploidy factor of 1.7 were estimated to change their karyotype at a rate of about 3% per generation, compared with 1.8% for cells with a ploidy factor of 0.95. Because the background noise of karyotyping is relatively high, the cells with low ploidy factor may be more stable than our method suggests. The karyotype instability of human colon cancer cell lines, recently analyzed by Lengnauer et al. [Lengnauer, C., Kinzler, K. W. & Vogelstein, B. (1997) Nature (London) 386, 623–627], also corresponds exactly to their degree of aneuploidy. We conclude that aneuploidy is sufficient to explain genetic instability and the resulting karyotypic and phenotypic heterogeneity of cancer cells, independent of gene mutation. Because aneuploidy has also been proposed to cause cancer, our hypothesis offers a common, unique mechanism of altering and simultaneously destabilizing normal cellular phenotypes.

Elections, information aggregation, and strategic voting

A central role of elections is the aggregation of information dispersed within a population. This article surveys recent work on elections as mechanisms for aggregating information and on the incentives for voters to vote strategically in such elections.

The rate constant of photoinhibition, measured in lincomycin-treated leaves, is directly proportional to light intensity.

Pumpkin leaves grown under high light (500-700 micromol of photons m-2.s-1) were illuminated under photon flux densities ranging from 6.5 to 1500 micromol.m-2.s-1 in the presence of lincomycin, an inhibitor of chloroplast protein synthesis. The illumination at all light intensities caused photoinhibition, measured as a decrease in the ratio of variable to maximum fluorescence. Loss of photosystem II (PSII) electron transfer activity correlated with the decrease in the fluorescence ratio. The rate constant of photoinhibition, determined from first-order fits, was directly proportional to photon flux density at all light intensities studied. The fluorescence ratio did not decrease if the leaves were illuminated in low light in the absence of lincomycin or incubated in darkness in the presence of lincomycin. The constancy of the quantum yield of photoinhibition under different photon flux densities strongly suggests that photoinhibition in vivo occurs by one dominant mechanism under all light intensities. This mechanism probably is not the acceptor side mechanism characterized in the anaerobic case in vitro. Furthermore, there was an excellent correlation between the loss of PSII activity and the loss of the D1 protein from thylakoid membranes under low light. At low light, photoinhibition occurs so slowly that inactive PSII centers with the D1 protein waiting to be degraded do not accumulate. The kinetic agreement between D1 protein degradation and the inactivation of PSII indicates that the turnover of the D1 protein depends on photoinhibition under both low and high light.

Electoral Systems and Ethnic Conciliation: A Structured, Focused Analysis of Vote-Pooling in Northern Ireland Elections 1998-2011

This research project examines the role of electoral system rules in affecting the extent of conciliatory behavior and cross-ethnic coalition making in Northern Ireland. It focuses on the role of the Single Transferable Vote (STV) electoral system in shaping party and voter incentives in a post-conflict divided society. The research uses a structured, focused comparison of the four electoral cycles since the Belfast Agreement of 1998. This enables a systematic examination of each electoral cycle using a common set of criteria focused on conciliation and cross-ethnic coalition making. Whilst preference voting is assumed to benefit moderate candidates, in Northern Ireland centrist and multi-ethnic parties outside of the dominant ethnic communities have received little electoral success. In Northern Ireland the primary effect of STV has not been to encourage inter-communal voting but to facilitate intra-community and intra-party moderation. STV has encouraged the moderation of the historically extreme political parties in each of the ethnic bloc. Patterns across electoral cycles suggest that party elites from the Democratic Unionist Party (DUP) and Sinn Fein have moderated their policy positions due to the electoral system rules. Therefore they have pursued lower-preference votes from within their ethnic bloc but in doing so have marginalized parties of a multi-ethnic or non-ethnic orientation.