Tests for Comparing Mark-Specific Hazards and Cumulative Incidence Functions

It is of interest in some applications to determine whether there is a relationship between a hazard rate function (or a cumulative incidence function) and a mark variable which is only observed at uncensored failure times. We develop nonparametric tests for this problem when the mark variable is continuous. Tests are developed for the null hypothesis that the mark-specific hazard rate is independent of the mark versus ordered and two-sided alternatives expressed in terms of mark-specific hazard functions and mark-specific cumulative incidence functions. The test statistics are based on functionals of a bivariate test process equal to a weighted average of differences between a Nelson--Aalen-type estimator of the mark-specific cumulative hazard function and a nonparametric estimator of this function under the null hypothesis. The weight function in the test process can be chosen so that the test statistics are asymptotically distribution-free.Asymptotically correct critical values are obtained through a simple simulation procedure. The testing procedures are shown to perform well in numerical studies, and are illustrated with an AIDS clinical trial example. Specifically, the tests are used to assess if the instantaneous or absolute risk of treatment failure depends on the amount of accumulation of drug resistance mutations in a subject's HIV virus. This assessment helps guide development of anti-HIV therapies that surmount the problem of drug resistance.

Lung-function tests in neonates and infants with chronic lung disease: tidal breathing and respiratory control

This paper is the fourth in a series of reviews that will summarize available data and critically discuss the potential role of lung-function testing in infants with acute neonatal respiratory disorders and chronic lung disease of infancy. The current paper addresses information derived from tidal breathing measurements within the framework outlined in the introductory paper of this series, with particular reference to how these measurements inform on control of breathing. Infants with acute and chronic respiratory illness demonstrate differences in tidal breathing and its control that are of clinical consequence and can be measured objectively. The increased incidence of significant apnea in preterm infants and infants with chronic lung disease, together with the reportedly increased risk of sudden unexplained death within the latter group, suggests that control of breathing is affected by both maturation and disease. Clinical observations are supported by formal comparison of tidal breathing parameters and control of breathing indices in the research setting.

A NOVEL AND SIMPLE RULE OF THUMB FOR MULTIPLICITY CONTROL IN EQUIVALENCE TESTING USING TWO ONE-SIDED TESTS

Equivalence testing is growing in use in scientific research outside of its traditional role in the drug approval process. Largely due to its ease of use and recommendation from the United States Food and Drug Administration guidance, the most common statistical method for testing (bio)equivalence is the two one-sided tests procedure (TOST). Like classical point-null hypothesis testing, TOST is subject to multiplicity concerns as more comparisons are made. In this manuscript, a condition that bounds the family-wise error rate (FWER) using TOST is given. This condition then leads to a simple solution for controlling the FWER. Specifically, we demonstrate that if all pairwise comparisons of k independent groups are being evaluated for equivalence, then simply scaling the nominal Type I error rate down by (k - 1) is sufficient to maintain the family-wise error rate at the desired value or less. The resulting rule is much less conservative than the equally simple Bonferroni correction. An example of equivalence testing in a non drug-development setting is given.

RANDOM EFFECTS MODELS IN A META-ANALYSIS OF THE ACCURACY OF DIAGNOSTIC TESTS WITHIN A GOLD STANDARD IN THE PRESENCE OF MISSING DATA

In evaluating the accuracy of diagnosis tests, it is common to apply two imperfect tests jointly or sequentially to a study population. In a recent meta-analysis of the accuracy of microsatellite instability testing (MSI) and traditional mutation analysis (MUT) in predicting germline mutations of the mismatch repair (MMR) genes, a Bayesian approach (Chen, Watson, and Parmigiani 2005) was proposed to handle missing data resulting from partial testing and the lack of a gold standard. In this paper, we demonstrate an improved estimation of the sensitivities and specificities of MSI and MUT by using a nonlinear mixed model and a Bayesian hierarchical model, both of which account for the heterogeneity across studies through study-specific random effects. The methods can be used to estimate the accuracy of two imperfect diagnostic tests in other meta-analyses when the prevalence of disease, the sensitivities and/or the specificities of diagnostic tests are heterogeneous among studies. Furthermore, simulation studies have demonstrated the importance of carefully selecting appropriate random effects on the estimation of diagnostic accuracy measurements in this scenario.

A review identifies and classifies reasons for ordering diagnostic tests

OBJECTIVE: To consider the reasons and context for test ordering by doctors when faced with an undiagnosed complaint in primary or secondary care. STUDY DESIGN AND SETTING: We reviewed any study of any design that discussed factors that may affect a doctor's decision to order a test. Articles were located through searches of electronic databases, authors' files on diagnostic methodology, and reference lists of relevant studies. We extracted data on: study design, type of analysis, setting, topic area, and any factors reported to influence test ordering. RESULTS: We included 37 studies. We carried out a thematic analysis to synthesize data. Five key groupings arose from this process: diagnostic factors, therapeutic and prognostic factors, patient-related factors, doctor-related factors, and policy and organization-related factors. To illustrate how the various factors identified may influence test ordering we considered the symptom low back pain and the diagnosis multiple sclerosis as examples. CONCLUSIONS: A wide variety of factors influence a doctor's decision to order a test. These are integral to understanding diagnosis in clinical practice. Traditional diagnostic accuracy studies should be supplemented with research into the broader context in which doctors perform their work.

Expression of TCL1 in hematologic disorders

OBJECTIVE: TCL1, MTCP1 and TCL1b are three members of a new family of oncogenes that are expressed in T cell leukemias of ataxia telangiectasia patients (T-PLL, T-CLL). TCL1 is located at 14q32.1 and activated by juxtaposition to the alpha/delta-locus at 14q11 or beta-locus at 7q35 of the T cell receptor during the reciprocal translocations t(14;14)(q11;q32), t(7;14)(q35;q32), or inversion inv(14)(q11;q32). TCL1 encodes a predominantly cytoplasmic protein of 114 aa (14 kD) of unknown function. Recent studies suggest that TCL1 promotes cell survival rather than stimulating cell proliferation, as previously proposed. METHODS: In an attempt to clarify the contexts in which TCL1 is expressed, we investigated TCL1 expression in 114 lymphoma and leukemia patients by Northern blot, RT-PCR and immunohistochemistry. RESULTS: TCL1 expression is restricted to lymphoid cells, and is found in neoplastic (T and B cell neoplasms, and Hodgkin's disease) and nonneoplastic proliferations (reactive lesions). Out of 114 cases, 18 neoplasms of myeloid and 4 cases of epithelial origin were TCL1-negative. In lesions of the lymphoid system, both low- and high-grade lymphomas were found to express TCL1. CONCLUSIONS: We propose that TCL1 expression especially in high-grade B cell non-Hodgkin's lymphomas might interfere with B cell differentiation and promote the transition from low- to high-grade lymphoma.

Confidence Bands for Convex Median Curves using Sign-Tests

Suppose that one observes pairs (x1,Y1), (x2,Y2), ..., (xn,Yn), where x1 < x2 < ... < xn are fixed numbers while Y1, Y2, ..., Yn are independent random variables with unknown distributions. The only assumption is that Median(Yi) = f(xi) for some unknown convex or concave function f. We present a confidence band for this regression function f using suitable multiscale sign tests. While the exact computation of this band seems to require O(n4) steps, good approximations can be obtained in O(n2) steps. In addition the confidence band is shown to have desirable asymptotic properties as the sample size n tends to infinity.

Risk factors for invasive aspergillosis in neutropenic patients with hematologic malignancies

Risk factors for invasive aspergillosis (IA) are incompletely identified and may undergo changes due to differences in medical practice. A cohort of 189 consecutive, adult patients with neutropenia hospitalized in the hemato-oncology ward of the University hospital Berne between 1995 and 1999 were included in a retrospective study to assess risk factors for IA. In total, 45 IA cases (nine proven, three probable, 33 possible), 11 patients with refractory fever and 133 controls were analyzed. IA cases had more often acute leukemia or myelodysplastic syndrome (MDS) (88 vs 38%, P < 0.001) and a longer duration of neutropenia (mean 20.6 vs 9.9 days, P < 0.001). They also had fewer neutropenic episodes during the preceding 6 months (mean 0.42 vs 1.03, P < 0.001), that is, confirmed (82%) and probable (73%) IA occurred most often during the induction cycle. A short time interval ( < or = 14 days) between neutropenic episodes increased the risk of IA four-fold (P = 0.06). Bacteremia, however, was not related to the number of preceding neutropenic episodes. Therefore, neutropenic patients with leukemia or MDS have the highest risk of IA. The risk is highest during the first induction cycle of treatment and increases with short-time intervals between treatment cycles.

Regional structural characteristics of bovine periodontal ligament samples and their suitability for biomechanical tests

Mechanical testing of the periodontal ligament requires a practical experimental model. Bovine teeth are advantageous in terms of size and availability, but information is lacking as to the anatomy and histology of their periodontium. The aim of this study, therefore, was to characterize the anatomy and histology of the attachment apparatus in fully erupted bovine mandibular first molars. A total of 13 teeth were processed for the production of undecalcified ground sections and decalcified semi-thin sections, for NaOH maceration, and for polarized light microscopy. Histomorphometric measurements relevant to the mechanical behavior of the periodontal ligament included width, number, size and area fraction of blood vessels and fractal analysis of the two hard-soft tissue interfaces. The histological and histomorphometric analyses were performed at four different root depths and at six circumferential locations around the distal and mesial roots. The variety of techniques applied provided a comprehensive view of the tissue architecture of the bovine periodontal ligament. Marked regional variations were observed in width, surface geometry of the two bordering hard tissues (cementum and alveolar bone), structural organization of the principal periodontal ligament connective tissue fibers, size, number and numerical density of blood vessels in the periodontal ligament. No predictable pattern was observed, except for a statistically significant increase in the area fraction of blood vessels from apical to coronal. The periodontal ligament width was up to three times wider in bovine teeth than in human teeth. The fractal analyses were in agreement with the histological observations showing frequent signs of remodeling activity in the alveolar bone - a finding which may be related to the magnitude and direction of occlusal forces in ruminants. Although samples from the apical root portion are not suitable for biomechanical testing, all other levels in the buccal and lingual aspects of the mesial and distal roots may be considered. The bucco-mesial aspect of the distal root appears to be the most suitable location.