Editorial Year 2004 – Report Another successful year has passed by and the journal continues to enjoy the status of most preferred among various groups of audience – from applied mathematicians/statisticians to biologists to agricul- turists. We have published 66 original papers in 2004. At a first glance, these seem to be rather heterogeneous reflecting the wide range of biometrical applications and the corresponding statistical methodology covered by Biometrical Journal and, indeed, we did not attempt a structure by topic so far. However, there are certain areas of methodological research and of applications that are very prominent in the 2004 issues of the journal such as categorical and count data, ordinal data, repeated measure and longitudinal data, methods for survival analysis or, more generally, event history data, clinical trial methodology, in particular adaptive and sequential designs, genetics and genomics, statis- tical methods in epidemiology, and finally, biology, ecology, and pharmacology. We are pleased to note that the following papers (in the respective order) were the ten most-down- loaded articles in the year 2004 from the website of the journal: Dahmen and Ziegler (2004), Munzel and Langer (2004), Stare and O’Quigley (2004), Lin and Wang (2004), Posch et al. (2004), Singer, Poleto and Rosa (2004), Feng and Kelly (2004), Park and Lee (2004), Overall and Tonidandel (2004) and Benda et al. (2004). Below we briefly mention and summarize some of the papers including the most downloaded ones which appeared in 2004 in order to help the readers to have an overview of the articles categorized under the most frequent special topics and areas. This will also aid in an easy looking-up of these papers. Concerning the range of applications, clinical trial methodology is still a topic of major relevance among the 2004 papers. The paper by Dahmen and Ziegler (2004) discusses the applicability of GEE as primary analysis in controlled clinical trials. Recommendations of how GEE should be used in therapeutic studies for testing statistical hypotheses are derived from theoretical results in the literature. For adaptive or sequential designs, Posch et al. (2004) consider the conditional rejection probability of the one-sided, one-sample t-test and give several proposals of how to implement de- sign adaptions in this case. The paper by Feng and Kelly (2004) extends the Model-Free Test (MFT) of Laska et al. (1994) to assess synergy in three or more drug combinations, and examines the effects of r different combination dosages and different dose-response functions on the power of the MFT. Additionally the power of the MFT is compared to the more traditional marginal dose- response curve method. The paper by Benda et al. (2004) provides sample size calculations needed for confidence interval estimation of the success probability measured by the Pearl Index in the clinical investigation of steroid contraceptives in women. The underlying model is discussed, and definitions and formulae are given for the assumption of a Poisson model. The necessary total expo- sure time is calculated as a function of a given true pregnancy rate. Liu et al. (2004) address the problem of estimating the treatment effect after stopping in a group sequential-trial while adjust- ments for binary covariates in clinical trials are discussed by Berger (2004). Finally, Andr�s and Herranz Tejedor (2004) present new tests for equivalence and for substantial difference in clinical trials settings. The authors demonstrate that their procedures are more powerful than the tests based on confidence intervals. With regard to methodological developments, analysis of categorical and binary data is well represented among the 2004 papers. Unconditional exact test procedures for the analysis of incomplete paired data are proposed by Tang and Tang (2004) and are compared with the asymptotic procedure by Choi and Stablein (1982). Kang and Kim (2004) provide power comparisons of three exact tests, namely Fisher’s test, the c2-test, and the likelihood ratio test. Lee and Qaqish (2004) address the problem of testing the goodness of fit of models for marginal probabilities estimated by GEEs in models for correlated binary data while Cohen et al. (2004) suggest a new test of the null hypothesis of the equality of the odds ratios in K ordered 2� 2 contingency tables. Biometrical Journal 47 (2005) 2, 113–118 DOI: 10.1002/bimj.200510105 # 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Several papers discuss methods for count data. Under the Poisson model, Bayesian methods for count data are presented by Stamey et al. (2004). A test for detecting zero inflation in Poisson models is suggested by Ugarte et al. (2004) while Dagne (2004) considers hierarchical Bayesian models for zero inflated count data incorporating both overdispersion and excess zeros. The area of ordinal variables is also addressed in several papers. The paper by Munzel and Langer (2004) discusses the rank approach for the analysis of ordered categorical data and compares it with the so-called ridit approach considered by Bross (1958). Singer et al. (2004) compared parametric and nonparametric methods for the analysis of repeated measures designs involving ordered categorical data. In particular, the rank method mixed model approach developed by Akritas and Brunner (1997) and further investigated by Brunner and Langer (2000) and by Shah and Madden (2004) is discussed in detail and compared with the na�ve parametric approach. Lui et al. (2004) develop two test statistics for testing equality of the diagnostic accuracy between two procedures for paired-samples ordinal data. Berger et al. (2004) propose new methods for adjusting for ordinal covariates without treating them as nominal or continuous. Repeated measurements and longitudinal data in various settings are discussed in many papers. Park and Lee (2004) propose simple residual plots for the investigation of the goodness of model fit for repeated measures data. The robustness of GEE tests to deviations from the assumption on the underlying covariance structure of the error term is investigated by Overall and Tonidandel (2004) while Wu et al. (2004) compare linear, nonlinear and semiparametric mixed-effect models for estimat- ing the decay rates in viral dynamic models. Wernecke et al. (2004) compare the random subject effect approach (Azzalini, 1995) with the cross-validated feature selection (Wernecke, 2002) in discri- minant analysis. In particular, the proposed method is useful in comparing several models simulta- neously. The impact of attrition and dropouts on the power of longitudinal studies is investigated by DuBois Bowman (2004) and two new methods for predicting the power are presented. A general marginal model for the analysis of percentage data, for example, pre-post percent change data in a longitudinal set-up, is considered by Song et al. (2004) extending the marginal simplex model by Song and Tan (2000) to the heteroscedastic case. Sample size calculations for several samples designs in repeated measurements studies are provided by Jung and Ahn (2004) extending the method of Jung and Ahn (2003) for the two-sample case. Many papers from the area of statistical methods for survival or, more generally, event history data were also published in 2004 and we will discuss only some of these including the most-down- loaded ones here. Stare and O’Quigley (2004) address the goodness of fit problem for proportional hazards regression using an equivalence between proportional hazards models with frailty terms and non-proportional hazards models in which regression effects are not constant in time (O’Quigley and Stare, 2002) while Lin and Wang (2004) derive a new test for comparing the overall homogeneity of several survival curves. The authors demonstrate that in the case of crossing survival curves this test has a higher power than commonly used tests. For doubly censored failure time data a method for testing the assumption of independence of the occurrence of the initial event is presented by Sun et al. (2004). Sæbø and Almøy (2004) present a method for fitting parametric models to complex hazard rates in survival data. The application of time-varying frailty models to clustered failure time data is discussed by Wintrebert et al. (2004). Statistical developments for epidemiological research, in general, genetics and genomics, biol- ogy, ecology and pharmacology still deserve a more prominent place in the journal. We do not list here the papers which appeared in these areas in 2004. Our desire is to have more articles in these areas and we would like to invite authors to submit high quality manucripts dealing with such topics. We would appreciate to share the experiences of the readers with the application of the methods proposed and discussed in the 2004-year papers published in Biometrical Journal. In order to make the journal more ‘alive’ we also invite our readers, authors and referees to send their opinions on papers published in the journal, in general and the above-listed papers, in particular, as ‘Letters to the Editors’. 114 Brunner and Schumacher: Editorial # 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Submissions of Manuscripts The journal started the year with a new Editorial Board and re-defined aims and scope to emphasise more on the applied nature of articles intended to be published. The year also saw complete electro- nic handling of manuscripts, thus aiding faster editorial process, with all manuscripts submitted via e-mail. We received 208 new manuscript submissions in 2004, of which ten were submitted for the special issue on “Therapeutic Equivalence” handled by the Guest Editors Axel Munk and Hans-J. Trampisch. Figure 1 shows the absolute frequencies of all the 208 incoming manuscripts over the twelve-month period. As of January 31, 2005, of the 198 regular manuscripts, there were 118 manuscripts rejected, 42 accepted, 22 under first review and 26 under second review. The Kaplan-Meier curves of the first review time (day of submission until either receipt of the first report of the Associate Editor or the decision on immediate rejection by the Editors) are displayed in Figure 2. The solid survival curve includes all manuscripts submitted in 2004, while the dashed curve refers to the first review time where the manuscripts not complying with the new aims and scope of the Journal (immediate rejec- tions) are excluded. New Aims and Scope All manuscripts which did not meet the new aims and scope of the Journal, and for this reason were not acceptable for publication, were rejected by the Editorial Board within about a week of their receipt. In this connection we would like to point out that each manuscript in which a new methodol- ogy is displayed, an example from Biometry should be discussed which motivates the theoretical derivations. Ideally, this example should be analyzed using the new method and the results should be compared to those from other existing methods and discussed accordingly. Review Times From Figure 2 it is seen that the median review time is about 7 weeks in total and 11 weeks if the immediate rejections are excluded. For the benefit of the authors submitting their manuscripts and a faster transfer of research findings to the readers, the Editors are striving best to shorten the length of review times. In many cases, we were successful in keeping the first round of reviews short. Unfortu- Biometrical Journal 47 (2005) 2 115 1 2 3 4 5 6 7 8 9 10 11 12 Month 0 5 10 15 20 25 10 21 20 24 16 22 18 20 13 16 14 14 Figure 1 Frequency of manuscripts arrival for each month in 2004. # 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim nately, we were not able to achieve this for every paper as we are dependant on our referees’ avail- ability of time to review papers apart from their professional activities. We consider it very important to have the manuscripts submitted to Biometrical Journal reviewed thoroughly and extensively; and thus, sometimes this takes time. For this reason, we ask our authors for their understanding regarding the review process. From the Editors’ side all efforts are made that the submitted manuscripts are handled as quickly as possible. We also ask our authors to submit manuscripts of reasonable length; lengthy papers are not encouraged. We would like to take this opportunity to thank all our Associate Editors and referees for their time and effort which they spent for running Biometrical Journal. Discussion Papers and Letters to the Editors In the future, we intend to publish more articles under special topics and also discussion papers are planned, especially on topics of high practical relevance and controversy. In particular, we would welcome letters to the Editors and invite our readers, authors and referees to submit their opinions and comments to all papers published in Biometrical Journal. Further Plans The publisher intends to provide Early Views for all accepted manuscripts on the Journal’s homepage. This means that these manuscripts will be considered as published and they can be refered to by their DOI-number, until actual printing occurs. We are also actively considering to introduce a system of electronic submission of manuscripts through a web-interface. We are looking forward to working with more authors presenting valuable and high-quality manu- scripts in the year 2005. As always, we are striving our best to keep review times shorter and widen our readership base. Edgar Brunner, G�ttingen Martin Schumacher, Freiburg References Akritas, M. G. and Brunner, E. (1997). A unified approach to rank tests for mixed models. Journal of Statistical Planning and Inference 61, 249–277. Andr�s, A. M. and Herranz Tejedor, I. (2004). Asymptotical tests on the equivalence, substantial difference and non-inferiority problems with two proportions. Biometrical Journal 46, 305–319. 116 Brunner and Schumacher: Editorial 0 5 10 15 20 25 30 0. 0 0. 2 0. 4 0. 6 0. 8 1. 0 First Review Time [Weeks] All manuscripts All manuscripts except immediately rejected ones Figure 2 Kaplan-Meier curves for all manu- scripts including and excluding immediately re- jected cases. # 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Azzalini, A. (1995). Topics in Repeated Measurements Analysis: ANOVA with Correlated Data. Technical report, University of Padua, Padua. Benda, N., Gerlinger, C., van der Meulen, E. A., and Endrikat, J. (2004). Sample size calculation for clinical studies on the efficacy of a new contraceptive method. Biometrical Journal 46, 141–150. Berger, V. W. (2004). Valid adjustment of randomized comparisons for binary covariates. Biometrical Journal 46, 589–594. Berger, V. W., Zhou, Y. Y., Ivanova, A., and Tremmel, L. (2004). Adjusting for ordinal covariates by inducing a partial ordering. Biometrical Journal 46, 48–55. Bross, I. D. J. (1958). How to use ridit analysis. Biometrics 14, 18–38. Brunner, E. and Langer, F. (2000). Nonparametric analysis of ordered categorical data in designs with longitudinal observations and small sample sizes. Biometrical Journal 42, 663–675. Choi, S. C. and Stablein, D. M. (1982). Practical tests for comparing two proportions with incomplete data. Applied Statistics 31, 256–262. Cohen, A., Kolassa, J., and Sackrowitz H. B. (2004). A test for homogeneity of odds ratios in ordered 2� 2 tables. Biometrical Journal 46, 633–641. Dagne, G. A. (2004). Hierarchical Bayesian analysis of correlated zero-inflated count data. Biometrical Journal 46, 653–663. Dahmen, G. and Ziegler, A. (2004). Generalized estimating equations in controlled clinical trials: Hypotheses testing. Biometrical Journal 46, 214–232. DuBois Bowman, F. (2004). Predicting power for longitudinal studies with attrition. Biometrical Journal 46, 453– 459. Feng, P. and Kelly, C. (2004). An extension of the model free test to test synergy in multiple drug combinations. Biometrical Journal 46, 293–304. Jung, S. H. and Ahn, C. (2003). Sample size estimation for GEE method for comparing slopes in repeated meas- urements data. Statistics in Medicine 22, 1305–1315. Jung, S.-H. and Ahn, C. (2004). K-Sample test and sample size calculation for comparing slopes in data with repeated measurements. Biometrical Journal 46, 554–564. Kang, S.-H. and Kim, S.-J. (2004). A comparison of the three conditional exact tests in two-way contingency tables using the unconditional exact power. Biometrical Journal 46, 320–330. Laska, E. M., Meisner, M., and Seigel, C. (1994). Simple designs and model-free tests for synergy. Biometrics 50, 834–841. Lee, J.-H. and Qaqish, B. F. (2004). Modified GEE and goodness of the marginal fit (GOMF) test with correlated binary responses for contingency tables. Biometrical Journal 46, 675–686. Lin, X. and Wang, H. (2004). A new testing approach for comparing the oerall homogeneity of survival curves. Biometrical Journal 46, 489–496. Liu, A., Troendle, J. F., Yu, K. F., and Yuan, V. W. (2004). Conditional maximum likelihood estimation following a group sequential test. Biometrical Journal 46, 760–768. Lui, K.-J., Zhou, X.-H., and Lin, C.-D. (2004). Testing equality between two diagnostic procedures in paired- sample ordinal data. Biometrical Journal 46, 642–652. Munzel, U. and Langer, F. (2004). A global view on parametric and nonparametric approaches to the analysis of ordered categorical data. Biometrical Journal 46, 7–18. O’Quigley, J. and Stare, J. (2002). Proportional Hazards Models with Frailties and Random Effects. Statistics in Medicine 21, 3219–3233. Overall, J. E. and Tonidandel, S. (2004). Robustness of generalized estimating equation (GEE) tests of signifi- cance against misspecification of the error structure model. Biometrical Journal 46, 203–213. Park, T. and Lee, S.-Y. (2004). Model diagnostic plots for repeated measures data. Biometrical Journal 46, 441– 452. Posch, M., Timmesfeld, N., K�nig, F., and M�ller, H.-H. (2004). Conditional rejection probabilities of student’s t-test and design adaptations. Biometrical Journal 46, 389–403. Sæbø, S. and Almøy, T. (2004). Composite hazard functions reflecting competing latent processes. Biometrical Journal 46, 103–113. Shah, D. and Madden, L. V. (2004). Nonparametric analysis of ordinal data in designed factorial experiments. Phytopathology 94, 33–43. Singer, J. M., Poleto, F. Z., and Rosa, P. (2004). Parametric and nonparametric analyses of repeated ordinal categorical data. Biometrical Journal 46, 460–473. Song, P. X.-K., Qiu, Z., and Tan M. (2004). Modelling heterogeneous dispersion in marginal models for longi- tudinal proportional data. Biometrical Journal 46, 540–553. Biometrical Journal 47 (2005) 2 117 # 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim Song, P. X.-K. and Tan, M. (2000). Marginal models for longitudinal continuous proportional data. Biometrics 56, 496–502. Stamey, J. D., Young, D. M., and Bratcher, T. L. (2004). Bayesian predictive probability functions for count data that are subject to misclassification. Biometrical Journal 46, 5742–578. Stare, J. and O’Quigley, J. (2004). Fit and frailties in proportional hazards regression. Biometrical Journal 46, 157–164. Sun, J., Lim, H.-J. and Zhao, X. (2004). An independence test for doubly censored failure time data. Biometrical Journal 46, 503–511. Tang, M.-L. and Tang, N.-S. (2004). Exact tests for comparing two paired proportions with incomplete data. Biometrical Journal 46, 72–82. Ugarte, M. D., Ib�nez, B., and Militino, A. F. (2004). Testing for poisson zero inflation in disease mapping. Biometrical Journal 46, 526–539. Wernecke, K. D. (2002). A new approach to discriminant analysis with longitudinal data. In: W. Gaul and G. Ritter (eds.), Classification, Automation, and New Media, pp. 523–529. Springer, Berlin, Heidelberg, New York. Wernecke, K.-D., Kalb, G., Schink, T., and Wegner, B. (2004). A mixed model approach to discriminant analysis with longitudinal data. Biometrical Journal 46, 246–254. Wintrebert, C. M. A., Putter, H., Zwinderman, A. H., and Houwelingen van, J. C. (2004). Centre-effect on survival after bone marrow transplantation: Application of time-dependent frailty models. Biometrical Journal 46, 512–525. Wu, H., Zhao, C., and Liang, H. (2004). Comparison of linear, nonlinear and semiparametric mixed-effects models for estimating HIV dynamic parameters. Biometrical Journal 46, 233–245. 118 Brunner and Schumacher: Editorial # 2005 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim