Resampling-based methods for the analysis of multiple endpoints in clinical trials

We consider multivariate tests for comparing two treatments with multiple endpoints. The test decision is drawn from the simultaneous consideration of the univariate tests for the single endpoints. A general class of these tests, called cut-off tests, can be given, which, however, can lead to highly conservative procedures because the dependencies among the endpoints are not taken into account. In applying resampling-based methods considerable improvements for these tests can be achieved. Resampling-based cut-off tests are proposed which are sensitive against a treatment difference in a single endpoint, in a subgroup of endpoints, or in all endpoints. The results of Monte Carlo simulations demonstrate that a remarkable gain in statistical power as compared to the crude simultaneous consideration can be reached. In particular, for the multivariate one-sided test situation the proposed tests can be recommended. As an example the application of the tests is demonstrated by data from a clinical trial.     

Long‐term tooth loss in periodontally compromised but treated patients according to the type of prosthodontic treatment. A retrospective study

If prosthodontic treatment is considered after periodontal therapy, the questions arise i) does prosthodontic treatment affect the treatment outcome of the dentition in general and ii) which type of prosthesis is related to best treatment outcome of abutment teeth? Our goal was to compare long‐term tooth loss after comprehensive periodontal therapy in patients with or without prosthodontic treatment. Ninety patients' charts with a total of 1937 teeth who had received comprehensive periodontal treatment 5–17 years ago by the same periodontist were retrospectively evaluated. Sixty‐five patients received fixed dental prostheses (FDP; n = 29) and/or removable partial dentures anchored with clips (RPDC; n = 25) or double crowns (RPDD; n = 25). Twenty‐five patients were also periodontally compromised but treated without prosthodontic treatment and served as a control group. A total of 317 teeth and 70 abutment teeth were lost during 9·7 ± 4·1 years of observation. Thereof, 273 teeth and 48 abutment teeth were lost due to periodontal reasons. Mean tooth loss amounted to 1·2 ± 1·5 (controls) and 4·4 ± 3·4 (partial dentures). Abutment tooth loss was 0·4 ± 1·1 (FDP), 1·0 ± 1·2 (RPDC) and 1·3 ± 1·0 (RPDD). Poisson regressions identified prosthodontic treatment, age, socio‐economic status, diabetes mellitus, mean initial bone loss and aggressive periodontitis as factors significantly contributing to tooth loss. Age, diabetes and non‐compliance contributed to abutment tooth loss. Not considering biomechanical factors, patients with prosthodontic reconstructions under long‐term supportive periodontal therapy were at higher risk for further tooth loss than patients without prostheses. Not only the type of partial denture but also the patient‐related risk factors were associated with abutment tooth loss.     

Development of the " core="" yellow="" flags="" index"="" (cyfi)="" as="" a="" brief="" instrument="" for="" the="" assessment="" of="" key="" psychological="" factors="" in="" patients="" undergoing="" spine="" surgery"="

European Spine Journal - Depression, anxiety, catastrophising, and fear-avoidance beliefs are key "yellow flags" (YFs) that predict a poor outcome in back patients. Most surgeons...     

Genetic risk for asthma, allergic rhinitis, and atopic dermatitis.

In order to explore the genetic risk of a child with a family history of allergies developing asthma, allergic rhinitis, or atopic dermatitis, questionnaires filled in by 6665 families were analysed. The data were collected in a population based cross sectional survey of 9-11 year old schoolchildren living in Munich and southern Bavaria. The relation between asthma, allergic rhinitis, and atopic dermatitis and the number of allergic first degree relatives, and the type of allergic disease was examined. Analyses were done separately for families with single or multiple allergic diseases. In families with one allergic parent the risk of the child developing asthma was increased by asthma in a parent, with an odds ratio (OR) of 2.6 (95% confidence interval 1.7 to 4.0) but not by parental allergic rhinitis with OR 1.0 (0.7 to 1.5) or atopic dermatitis, OR 1.0 (0.6 to 1.6). For allergic rhinitis the highest risk with OR 3.6 (2.9 to 4.6) was observed with allergic rhinitis of one parent, apparently lower for asthma of one parent, OR 2.5 (1.6 to 4.0) or atopic dermatitis, OR 1.7 (1.1 to 2.5). Children with parental atopic dermatitis had a high risk for atopic dermatitis, OR 3.4 (2.6 to 4.4), compared with children with parental asthma, OR 1.5 (1.0 to 2.2), or parental allergic rhinitis, OR 1.4 (1.1 to 1.8). Risk factors in families with combined allergies of two relatives (parents and siblings) were analysed separately for the different combinations. These results support the hypothesis that asthma, allergic rhinitis, and atopic dermatitis are multifactorial diseases brought about by various familial and environmental influences.     

A randomised,placebo-controlled study comparing two formulations of dimenhydrinate with respect to efficacy in motion sickness and sedation

It is known that sedation by H1 antihistaminic drugs can be reduced or avoided if slow release formulations are used for their administration, probably because of a slower increase of the drug concentration in plasma and brain. The aim of this study was to compare two different formulations of dimenhydrinate (CAS 523-87-5), a single fast release tablet with three chewing gums (divided dose principle), with regard to their efficacy in a motion sickness model and their detrimental effect on vigilance and central nervous system (CNS) performance. Caloric stimulation of the eardrum (air at 44 degrees C) was used to induce the symptoms of motion sickness in 24 symptomatic volunteers in a three-way cross-over design comparing three chewing gums (Superpep forte, chewed for 30 min each) containing 20 mg dimenthydrinate each with a 50 mg dimenhydrinate tablet and placebo. During caloric stimulation the following parameters were measured in order to compare efficacy: Quantitative analysis of sodium excretion by sweat (main target parameter), subjective well being (vertigo) by visual analogue scales (VAS) and frequency of binocular nystagmus by computer nystagmography. Unwanted effects on vigilance and CNS performance were measured by means of the N1-P2 peak to peak amplitudes of auditory evoked potentials (AEPs) as an objective, quantitative parameter of vigilance and the latency to correct responses and the number of correct responses (complex choice reaction task) in the oculodynamic test (ODT) as parameters of complex choice reaction ability. As a main efficacy result sodium excretion by sweat was markedly reduced by the chewing gums and by the tablet. The differences to placebo were highly significant (chewing gums vs. placebo p < 0.0001, tablet vs. placebo p < 0.0001). There was no relevant and no significant difference between both medications (p = 0.308). The secondary efficacy parameters, frequency of binocular nystagm and the VAS vertigo were markedly reduced by both medications in comparison to placebo, i.e. both medications were markedly effective. In both cases, however, this result failed statistical significance. The unwanted depressing effects on vigilance and CNS performance of the chewing gums were less pronounced than that of tablets. The N1/P2 peak-to-peak amplitudes of the AEPs were significantly reduced by both the chewing gums and the tablets. The effect of the tablets was, however, larger than that of the chewing gums. This highly significant (tablet vs. chewing gums, p = 0.0003) difference shows that the tablet had a larger depressing effect on vigilance (greater sedation). In line with this result, the number of correct responses in the ODT was markedly and significantly reduced by the tablet (p = 0.0027) but not significantly by the chewing gums (p = 0.8140). The difference between the tablet and the chewing gums was highly significant (p = 0.0052). The complex choice reaction time was markedly and nearly significantly (p = 0.0558) prolonged by the tablet whereas the chewing gums produced only a very small and insignificant prolongation. That the objective measurements of vigilance and CNS performance showed significantly larger detrimental effects of the tablet than of the chewing gums is probably a consequence of a faster increase of the dimenhydrinate concentration in the CNS after administration of the tablet in comparison to the divided dose principle of the chewing gums.     

Radiographic parameters as prognostic indicators for healing of class II furcation defects

OBJECTIVE: To evaluate radiographic measurements for use as prognostic indicators for healing of class II furcation defects following regenerative therapy. MATERIAL AND METHODS: In 17 patients (eight females), 33 class II furcation defects (mandibular buccal (n=10) and lingual (n=12), and maxillary buccal (n=11)) were treated using the barrier membrane technique. Twenty-six furcations were treated using a bioabsorbable membrane, while a nonresorbable membrane was used to treat the remaining seven furcation defects. Clinical parameters and standardized radiographs were obtained before as well as 6 and 24 months after therapy. All radiographs were digitized and evaluated by an examiner blinded to the clinical data. The following distances were measured: cemento-enamel junction line (CEJ-line) to alveolar crest (AC) at the furcation site (AC-CEJ line), CEJ-line to the furcation fornix (Fx-CEJ line), width of the furcation at the level of the AC (FW) as well as the distance from Fx to a straight line between the AC mesial and distal of the tooth (Fx-AC line). RESULTS: Statistically significant (p<0.001) horizontal attachment gains could be observed 6 and 24 months after therapy (6 months: 1.49+/-0.85 mm; 24 months: 1.14+/-0.91 mm). However, a small but statistically significant (p=0.031) attachment loss of 0.35 mm was observed between the 6 and 24 months examination. Multilevel regression analyses identified baseline probing depth (p=0.0017) and 3 of the radiographic distances as prognostic factors: Fx-CEJ line (p=0.014), FW (p=0.0535), Fx-AC line (p=0.0827). CONCLUSION: The analysis of presurgical radiographs may yield information on the success of the regenerative therapy of buccal and lingual class II furcation defects. A long root trunk, a wide furcation entrance and an Fx coronal to the AC have negative influences on the success of therapy. Further, a deep probing depth at the furcation site at baseline increases the likelihood for more favourable horizontal attachment gain in furcations.     

Zur Erklärungskraft des heutigen soziodemografischen Risikostrukturausgleichsmodells

Zusammenfassung  Eine wesentliche Voraussetzung für effizienzsteigernden regulierten Wettbewerb in sozialen Krankenversicherungssystemen ist die weitgehende Aufhebung von Selektionsanreizen und Antiselektionseffekten durch Risikoausgleichsmechanismen. Damit solche Mechanismen wirksam sind, ist es u. a. wichtig, dass individuelle Unterschiede im Risiko der Verursachung von Leistungsausgaben hinreichend genau abgebildet werden. Internationale Erfahrungen haben gezeigt, dass Ausgleichsmodelle mit Alter und Geschlecht als alleinige Risikoindikatoren nur einen recht geringen Teil der individuellen Leistungsausgaben erklären. Diese Erkenntnis bestätigt sich auch in der vorliegenden Arbeit an den Daten einer ostdeutschen Regionalkasse. Ein weiteres Ergebnis ist, dass die zusätzliche Verwendung eines Indikators für den Bezug einer Erwerbsminderungsrente die Erklärungskraft des deutschen Risikostrukturausgleichsmodells deutlich verbessert. Allerdings ist auch ein solches soziodemografisches Modell nicht in der Lage, Selektionsanreize ausreichend aufzuheben. Insbesondere bei Hochkostenfällen, BU/EU-Rentnern und Versicherten ab 60 Jahren kann mit beträchtlichen Abweichungen zwischen den mit diesen Ausgleichsparametern als risikobedingt veranschlagten und den tatsächlichen Ausgaben gerechnet werden.

Social inequalities and outcomes in type 2 diabetes in the German region of Augsburg. A cross-sectional survey.

Summary Objectives: To assess the association of socioeconomic position with health (care) outcomes in type 2 diabetes with a particular focus on glycaemic control. Methods: A cross-sectional survey in the region of Augsburg (Germany) on 373 men and women with type 2 diabetes, drawn from representative MONICA surveys and the myocardial infarct register. Analysis of association of socioeconomic position with HbA1c levels, cardiovascular risk factors and long-term macro- and microvascular diabetes complications using logistic regression models. Results: Glycaemic control, measured by HbA1c levels, is strongly associated with indicators of socioeconomic position favouring the better off. Comparison of the lowest with the highest socioeconomic group showed an odds ratio of 2.49 (95% CI: 1.22–5.07) for the MI register subgroup and 1.80 (95% CI: 0.80–4.06) for the survey subgroup for failure to achieve the recommended HbA1c target. This association could not be accounted for by differences across social groups in age, sex, diabetes duration, obesity, or physical activity. Conclusions: Social inequalities in glycaemic control do exist. This finding may indicate a level of diabetes care that is inappropriate to the need of socially disadvantaged groups. Submitted: 22 June 2005; Revised: 15 September 2006; Accepted: 13 March 2007     

Oxidative stress and inherited cataracts in mice

To determine whether an unbalanced redox state might accompany the development of particular inherited mouse cataracts, the lenticular content of oxidized glutathione (GSSG) and the activity of superoxide dismutase (SOD) were chosen as markers. For wild-type lenses, an enhanced GSSG content could be observed in females as compared to males. Such a sex effect could not be detected for the SOD activity. In the mutants, GSSG content in cataractous lenses was found to be enhanced in 2 of 7 cases; the increases in other mutants were not significant. Changes of the SOD activity were even less consistent and only a random correlation of GSSG content and SOD activity with cataractogenesis could be deduced.     

HEALTH EFFECTS OF SULFUR-RELATED ENVIRONMENTAL AIR POLLUTION. III. Nonspecific Respiratory Defense Capacities

Recently concern has been raised about health effects related to environmental sulfur and/or acidic aerosols. To assess long-term effects on respiratory lung function, 8 beagle dogs were exposed over a period of 13 mo for 16.5 h/day to 1.0 mum neutral sulfite aerosol with a particle associated sulfur (IV) concentration of 0.32 mg m-3 and for 6 h/ day to 1.1 mum acidic sulfate aerosol providing an hydrogen ion concentration of 15.2 mumol m-3 for inhalation. Prior to exposure the dogs were kept under clean air conditions for 16 mo to establish physiological baseline values for each dog. A second group of eight dogs (control) was kept for the entire study under clean air conditions. Nonspecific defense mechanisms in the airways and in the peripheral lung were studied during chronic exposure of the combination of neutral sulfur(IV) and acidic sulfur(VI) aerosols. No functional changes of tracheal mucus velocity were found, in agreement with unchanged morphometry of the airways. However, the exposure resulted in changes of several alveolar macrophage (AM) mediated particle clearance mechanisms: (1) Based on in vivo clearance analysis and cultured AM studies using moderately soluble cobalt oxide particles, intracellular particle dissolution was significantly reduced since phagolysosomal proton concentration was decreased. We deduce exposure-related malfunction of proton pumps bound to the phagolysosomal membrane as a result of an increase of cytosolic proton concentration. (2) Based on in vivo clearance analysis using insoluble polystyrene particles, AM-mediated particle transport from the lung periphery toward ciliated terminal bronchioli and further to the larynx was significantly reduced. Activation of epithelial type II cells at the entrance of alveoli was inferred from observed type II cell proliferation at those alveolar ridges and enhanced secretion of alkaline phosphatase in the fluid of bronchoalveolar lavages. As a result, hypersecretion of chemotactic mediators by activated type II cells at these loci led to the observed decrease of particle transport toward ciliated bronchioli. (3) Based on in vivo clearance analysis using insoluble polystyrene particles, particle transport from the alveolar epithelium into interstitial tissues was increased and (4) particle transport to the tracheobronchial lymph nodes was significantly enhanced. Particle transport into interstitial tissues is the most prominant clearance pathway from the canine alveolar epithelium. We conclude that the deteriorated particle transport toward ciliated terminal bronchioli resulted in an enhanced particle transport across the epithelial membrane into interstitial tissues and the lymphatic drainage. The observed alterations in alveolar macrophage-mediated clearance mechanisms during chronic exposure of these air pollutants indicate an increased risk of health.