PubMed Health⌕ Search

Biomedical subjects

Dankmar Böhning

Publications and source records attributed to Dankmar Böhning.

10 recordsLinked to original sources
PubMed2007

A comparison of three different models for estimating relative risk in meta-analysis of clinical trials under unobserved heterogeneity.

We focus on the comparison of three statistical models used to estimate the treatment effect in meta-analysis when individually pooled data are available. The models are two conventional models, namely a multi-level and a model based upon an approximate likelihood, and a newly developed model, the profile likelihood model which might be viewed as an extension of the Mantel-Haenszel approach. To exemplify these methods, we use results from a meta-analysis of 22 trials to prevent respiratory tract infections. We show that by using the multi-level approach, in the case of baseline heterogeneity, the number of clusters or components is considerably over-estimated. The approximate and profile likelihood method showed nearly the same pattern for the treatment effect distribution. To provide more evidence two simulation studies are accomplished. The profile likelihood can be considered as a clear alternative to the approximate likelihood model. In the case of strong baseline heterogeneity, the profile likelihood method shows superior behaviour when compared with the multi-level model.

Clinical Trials as Topic↗
PubMed2006

Comparison of two alternative methods for CD4+ T-cell determination (Coulter manual CD4 count and CyFlow) against standard dual platform flow cytometry in Uganda.

BACKGROUND: In this study we evaluated alternative CD4(+) T-cell counting methods in clients of a PMTCT Programme in rural Uganda. METHODS: The Coulter Manual CD4 Count method for CD4(+) T-cell enumeration (Cyto-Spheres) and an automated method (volumetric, single-platform flow cytometry; CyFlow) were compared with a standard, dual-platform flow cytometry protocol (DPFC, FACScan). RESULTS: Correlation and precision of agreement were higher for the CyFlow method (r = 0.929 and eta = 0.08) when compared to DPFC than for the Cyto-Spheres method (r = 0.725 and eta = 0.3). Multiple linear regression analysis showed that CD4(+) cell counts by the CyFlow method were a stronger predictor for results of DPFC than those of the Cyto-Spheres method (r(2) = 0.864 and r(2) = 0.552, respectively). When compared to DPFC the CyFlow method generated higher CD4(+) cell counts than the Cyto-Spheres method, as expressed by a higher median and mean difference (+70 and +90 cells for CyFlow, +28 and -1.4 cells for Cyto-Spheres). CONCLUSION: Both, the manual Cyto-Spheres method and the CyFlow method can be used for the enumeration of CD4(+) cells in resource-limited settings. Under supervised conditions, the CyFlow method produced results more consistent with the reference method than the Cyto-Spheres method.

Algorithms↗
PubMed2006

Evaluation of the cumulative evidence for freedom from BSE in birth cohorts.

Substantial resources are used for surveillance of bovine spongiform encephalopathy (BSE) despite an extremely low detection rate, especially in healthy slaughtered cattle. We have developed a method based on the geometric waiting time distribution to establish and update the statistical evidence for BSE-freedom for defined birth cohorts using continued surveillance data. The results suggest that currently (data included till September 2004) a birth cohort of Danish cattle born after March 1999 is free from BSE with probability (power) of 0.8746 or 0.8509, depending on the choice of a model for the diagnostic sensitivity. These results apply to an assumed design prevalence of 1 in 10,000 and account for prevalence heterogeneity. The age-dependent, diagnostic sensitivity for the detection of BSE has been identified as major determinant of the power. The incorporation of heterogeneity was deemed adequate on scientific grounds and led to improved power values. We propose our model as a decision tool for possible future modification of the BSE surveillance and discuss public health and international trade implications.

Animals↗
PubMed2006

Equivalence of truncated count mixture distributions and mixtures of truncated count distributions.

This article is about modeling count data with zero truncation. A parametric count density family is considered. The truncated mixture of densities from this family is different from the mixture of truncated densities from the same family. Whereas the former model is more natural to formulate and to interpret, the latter model is theoretically easier to treat. It is shown that for any mixing distribution leading to a truncated mixture, a (usually different) mixing distribution can be found so that the associated mixture of truncated densities equals the truncated mixture, and vice versa. This implies that the likelihood surfaces for both situations agree, and in this sense both models are equivalent. Zero-truncated count data models are used frequently in the capture-recapture setting to estimate population size, and it can be shown that the two Horvitz-Thompson estimators, associated with the two models, agree. In particular, it is possible to achieve strong results for mixtures of truncated Poisson densities, including reliable, global construction of the unique NPMLE (nonparametric maximum likelihood estimator) of the mixing distribution, implying a unique estimator for the population size. The benefit of these results lies in the fact that it is valid to work with the mixture of truncated count densities, which is less appealing for the practitioner but theoretically easier. Mixtures of truncated count densities form a convex linear model, for which a developed theory exists, including global maximum likelihood theory as well as algorithmic approaches. Once the problem has been solved in this class, it might readily be transformed back to the original problem by means of an explicitly given mapping. Applications of these ideas are given, particularly in the case of the truncated Poisson family.

Algorithms↗
PubMed2005

Revisiting proportion estimators.

Proportion estimators are quite frequently used in many application areas. The conventional proportion estimator (number of events divided by sample size) encounters a number of problems when the data are sparse as will be demonstrated in various settings. The problem of estimating its variance when sample sizes become small is rarely addressed in a satisfying framework. Specifically, we have in mind applications like the weighted risk difference in multicenter trials or stratifying risk ratio estimators (to adjust for potential confounders) in epidemiological studies. It is suggested to estimate p using the parametric family p(c) and p(1 - p) using p(c)(1 - p(c)), where p(c) = (X + c)/(n + 2c). We investigate the estimation problem of choosing c > or = 0 from various perspectives including minimizing the average mean squared error of p(c), average bias and average mean squared error of p(c)(1 - p(c)). The optimal value of c for minimizing the average mean squared error of p(c) is found to be independent of n and equals c = 1. The optimal value of c for minimizing the average mean squared error of p(c)(1 - p(c)) is found to be dependent of n with limiting value c = 0.833. This might justify to use a near-optimal value of c = 1 in practice which also turns out to be beneficial when constructing confidence intervals of the form p(c)+/-1.96 square root of np(c)(1 - p(c))/(n + 2c).

Bias↗
PubMed2004

Estimating the number of drug users in Bangkok 2001: a capture-recapture approach using repeated entries in one list.

BACKGROUND: Conventionally, capture-recapture techniques involving different lists such as police or hospitals are used for quantifying populations which are difficult to count, such as illicit drug user populations. Here, a novel approach is suggested based upon repeated entries in one list, which is less dependent on matching entries from different sources as in the conventional approach. METHODS: For this purpose, a population-based study was conducted that utilizes all data on treatment episodes of drug users from all 61 health treatment centers in the Bangkok metropolitan region to estimate the size of drug use in the Bangkok metropolitan region. The data stem from the drug treatment surveillance system of the Office of the Narcotics Control Board (ONCB) and cover the period from October 1 to December 31, 2001. Based upon the frequency of treatment episodes of each patient, a count distribution arose which could be modelled well by means of a Poisson mixture model. Using this count model, an estimate for the number of unobserved drug users could be constructed. RESULTS: From 11,222 drug users found during the period, 7063 (62.9%) were heroin users, 3346 (29.8%) metamphetamine users, and the remaining 813 (7.3%) distributed under 15 drug categories, none above 1%. The study concentrated on heroin and metamphetamine users who were predominantly male (96.2% for heroin and 91.8% for metamphetamine). Metamphetamine users were younger than heroin users (22.3 years 95% CI: 22.1-22.5 vs. 30.8 years 95% CI: 30.6-31.0). By using the truncated Poisson mixture model, an estimate of the unobserved frequency of drug users with zero treatment episodes could be constructed leading to an estimate of 11,296 (95% CI: 8,964-13,628) heroin users (completeness of identification: 38.42, 95% CI: 34.03-44.04%) and 32,105 (95% CI: 24,647-39,563) metamphetamine users (completeness of identification: 9.44, 95% CI: 7.79-11.97%) for the Bangkok metropolitan region. CONCLUSIONS: The proposed model showed excellent goodness-of-fit, unspecified for drug type and also if specified for the major drug types which allowed the prediction of the unobserved number of drug users in a realistic way, avoiding artefacts due to severe matching problems when using several, different sources. The technique is also easy to implement and can be used routinely to monitor drug user populations in space and time.

Adult↗
PubMed2004

A comparison of non-iterative and iterative estimators of heterogeneity variance for the standardized mortality ratio.

This paper continues work presented in Böhning et al. (2002b, Annals of the Institute of Statistical Mathematics 54, 827-839, henceforth BMSRB) where a class of non-iterative estimators of the variance of the heterogeneity distribution for the standardized mortality ratio was discussed. Here, these estimators are further investigated by means of a simulation study. In addition, iterative estimators including the Clayton-Kaldor procedure as well as the pseudo-maximum-likelihood (PML) approach are added in the comparison. Among all candidates, the PML estimator often has the smallest mean square error, followed by the non-iterative estimator where the weights are proportional to the external expected counts. This confirms the theoretical result in BMSRB in which an asymptotic efficiency could be proved for this estimator (in the class of non-iterative estimators considered). Surprisingly, the Clayton-Kaldor iterative estimator (often recommended and used by practitioners) performed poorly with respect to the MSE. Given the widespread use of these estimators in disease mapping, medical surveillance, meta-analysis and other areas of public health, the results of this study might be of considerable interest.

Berlin↗
PubMed2004

A mixture model application in disease mapping of malaria.

Disease mapping, a method for displaying the geographical distribution of disease occurrence, has received attention for more than 2 decades. Because traditional approaches to disease mapping have some deficiencies and disadvantages in presenting the geographical distribution of disease, the mixture model--as an alternative approach--overcomes some of these deficiencies and provides a clearer picture of the spatial risk structure. The purpose of this study was twofold: (1) to investigate the geographical distribution of malaria in Thailand during 1995, 1996, and 1997 by applying the mixture model to disease mapping, and (2) to investigate the dynamic nature of malaria in Thailand during the 3-year time frame by applying the space-time mixture model. Non-parametric maximum likelihood estimation was employed to estimate the parameters of both the mixture model and the space-time mixture model. Applying Bayes' theorem, the 76 provinces of Thailand were classified into component risk levels by the rate of malaria for each province. Malaria intensively occurred in 4 provinces on the Thai-Myanmar border and in 2 provinces on the Thai-Cambodian border. Of the 76 provinces studied, 10 showed an increasing trend over the 3-year period. A comparison of the map based on the mixture model with the map based on the traditional percentiles method indicates that the non-parametric mixture model removes random variability from the map and provides a clearer picture of the spatial risk structure. The advantage of the mixture model approach to disease mapping is the graphical visual presentation of the prevalence of disease. The space-time mixture model more adequately investigates the dynamic nature of disease than does the mixture model.

Bayes Theorem↗
PubMed2002

Some general points in estimating heterogeneity variance with the DerSimonian-Laird estimator.

In this paper we consider estimating heterogeneity variance with the DerSimonian-Laird (DSL) estimator as typically used in meta-analysis. In its general form the DSL estimator requires inverse population-averaged study-specific variances as weights, in which case the estimator is unbiased. It has become common practice, however, to use estimates of the study-specific variances instead of their population-averaged versions. This can lead to considerable bias. Simulations illustrate these findings.

Journal Article↗