Bivariate meta-analysis of predictive values of diagnostic tests can be an alternative to bivariate meta-analysis of sensitivity and specificity

Meta-analysis of predictive values is usually discouraged because these values are directly affected by disease prevalence, but sensitivity and specificity sometimes show substantial heterogeneity as well. We propose a bivariate random-effects logitnormal model for the meta-analysis of the positive predictive value (PPV) and negative predictive value (NPV) of diagnostic tests.

Dysphonic newborn cries allow prediction of their perceived meaning

Can adults attribute different meanings to the cries produced by the newborns on the basis of physical-acoustic cues in different communication conditions? In order to test this hypothesis, 20 females were asked to evaluate (according to four scales: anguish, anger, annoyance, care-seeking) 24 cries from 12 healthy full-term newborns (4-7h old): 6 newborns previously exposed to tactile communication (Group 1); 6 newborns without communication (Group 2). Annoyance category was not reliable and it was excluded from analyses. The 20 females attributed a higher rate of anger and anguish to the cries from Group 2, and a higher rate of care-seeking to the cries from Group 1. They attributed different meanings to the cries from Group 1, and undifferentiated meanings to the cries from Group 2. Consistent with bivariate analyses, Dysphonic Cry was the strongest predictor of anger/anguish. Although the Hyperphonic Cry was quantitatively not relevant, its absence was the first predictor for care-seeking.

Uniform approach to linear and nonlinear interrelation patterns in multivariate time series

Currently, a variety of linear and nonlinear measures is in use to investigate spatiotemporal interrelation patterns of multivariate time series. Whereas the former are by definition insensitive to nonlinear effects, the latter detect both nonlinear and linear interrelation. In the present contribution we employ a uniform surrogate-based approach, which is capable of disentangling interrelations that significantly exceed random effects and interrelations that significantly exceed linear correlation. The bivariate version of the proposed framework is explored using a simple model allowing for separate tuning of coupling and nonlinearity of interrelation. To demonstrate applicability of the approach to multivariate real-world time series we investigate resting state functional magnetic resonance imaging (rsfMRI) data of two healthy subjects as well as intracranial electroencephalograms (iEEG) of two epilepsy patients with focal onset seizures. The main findings are that for our rsfMRI data interrelations can be described by linear cross-correlation. Rejection of the null hypothesis of linear iEEG interrelation occurs predominantly for epileptogenic tissue as well as during epileptic seizures.

Analysis of the erosive effect of different dietary substances and medications

Excessive consumption of acidic drinks and foods contributes to tooth erosion. The aims of the present in vitro study were twofold: (1) to assess the erosive potential of different dietary substances and medications; (2) to determine the chemical properties with an impact on the erosive potential. We selected sixty agents: soft drinks, an energy drink, sports drinks, alcoholic drinks, juice, fruit, mineral water, yogurt, tea, coffee, salad dressing and medications. The erosive potential of the tested agents was quantified as the changes in surface hardness (ΔSH) of enamel specimens within the first 2 min (ΔSH2-0 = SH2 min - SHbaseline) and the second 2 min exposure (ΔSH4-2 = SH4 min - SH2 min). To characterise these agents, various chemical properties, e.g. pH, concentrations of Ca, Pi and F, titratable acidity to pH 7·0 and buffering capacity at the original pH value (β), as well as degree of saturation (pK - pI) with respect to hydroxyapatite (HAP) and fluorapatite (FAP), were determined. Erosive challenge caused a statistically significant reduction in SH for all agents except for coffee, some medications and alcoholic drinks, and non-flavoured mineral waters, teas and yogurts (P < 0·01). By multiple linear regression analysis, 52 % of the variation in ΔSH after 2 min and 61 % after 4 min immersion were explained by pH, β and concentrations of F and Ca (P < 0·05). pH was the variable with the highest impact in multiple regression and bivariate correlation analyses. Furthermore, a high bivariate correlation was also obtained between (pK - pI)HAP, (pK - pI)FAP and ΔSH.

Factors determining the success of radiofrequency denervation in lumbar facet joint pain: a prospective study

Background and aims Radiofrequency denervation (RF) of the lumbar facet joints has been shown to be effective in well-selected patients. However, long-term success varies between studies. We evaluated the influence of selected psychosocial and constitutional factors on the outcome of RF, expressed as the duration of pain relief. Methods This prospective observational study included 44 patients who received RF denervations at the University Hospital of Berne. Success was defined as at least 50% pain reduction 7–21 days, 6 months and 1 year after RF therapy. The Cox-regression analysis was performed to evaluate the influence of the following factors on the duration of success: age, sex, depression, work inability and previous surgery. Results Complete follow-up was available for 41 patients. The success rate 7–21 days after the denervation was 76%. It decreased to 32% at 6 months and to 22% at 1 year. The median success duration was 17 weeks (95% CI 10–26). The Cox-regression analysis showed a significant shorter duration of success for patients with depression (hazard ratio [HR] 2.97, 95% CI 1.32–6.65), previous surgery (HR 2.39, 95% CI 1.10–5.21) and number of treated joints (HR 1.95 for each increase in the number of joints, 95% CI 1.14–3.33). In bivariate analyses, only depression was kept to be significant. Conclusions Depression seems to be related with a short duration of success. Based on these findings, a comprehensive study is warranted to evaluate whether psychosocial factors have to be considered when recruiting patients for radiofrequency denervation.

Evaluation of radial distribution of cartilage degeneration and necessity of pre-contrast measurements using radial dGEMRIC in adults with acetabular dysplasia

Background The purpose of the present study was to investigate the radial distribution patterns of cartilage degeneration in dysplastic hips at different stages of secondary osteoarthritis (OA) by using radial delayed gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC), and to assess whether pre-contrast measurements are necessary. Methods Thirty-five hips in 21 subjects (mean age ± SD, 27.6 ± 10.8 years) with acetabular dysplasia (lateral CE angle < 25°) were studied. Severity of OA was assessed on radiographs using Tönnis grading. Pre- (T1pre) and post-contrast T1 (T1Gd) values were measured at 7 sub-regions on radial reformatted slices acquired from a 3-dimensional (3D) T1 mapping sequence using a 1.5 T MR scanner. Values of radial T1pre, T1Gd and ΔR1 (1/T1Gd - 1/T1pre) of subgroups with different severity of OA were compared to those of the subgroup without OA using nonparametric tests, and bivariate linear Pearson correlations between radial T1Gd and ΔR1 were analyzed for each subgroup. Results Compared to the subgroup without OA, the subgroup with mild OA was observed with a significant decrease in T1Gd in the anterosuperior to superior sub-regions (mean, 476 ~ 507 ms, p = 0.026 ~ 0.042) and a significant increase in ΔR1 in the anterosuperior to superoposterior and posterior sub-regions (mean, 0.93 ~ 1.37 s-1, p = 0.012 ~ 0.042). The subgroup with moderate to severe OA was observed with a significant overall decrease in T1Gd (mean, 404 ~ 452 ms, p = 0.001 ~ 0.020) and an increase in ΔR1 (mean, 1.17 ~1.69 s-1, p = 0.001 ~ 0.020). High correlations were observed between radial T1Gd and ΔR1 for all subgroups (r = −0.869 ~ −0.944, p < 0.001). Conclusions Radial dGEMRIC without pre-contrast measurements is useful for evaluating different patterns of cartilage degeneration in the entire hip joint of patients with hip dysplasia, particularly for those in early stages of secondary OA.

Genuine cross-correlations: Which surrogate based measure reproduces analytical results best?

The analysis of short segments of noise-contaminated, multivariate real world data constitutes a challenge. In this paper we compare several techniques of analysis, which are supposed to correctly extract the amount of genuine cross-correlations from a multivariate data set. In order to test for the quality of their performance we derive time series from a linear test model, which allows the analytical derivation of genuine correlations. We compare the numerical estimates of the four measures with the analytical results for different correlation pattern. In the bivariate case all but one measure performs similarly well. However, in the multivariate case measures based on the eigenvalues of the equal-time cross-correlation matrix do not extract exclusively information about the amount of genuine correlations, but they rather reflect the spatial organization of the correlation pattern. This may lead to failures when interpreting the numerical results as illustrated by an application to three electroencephalographic recordings of three patients suffering from pharmacoresistent epilepsy.

Does a pleiotropic gene explain deafness and blue irises in white cats?

The prevalence of deafness is high in cat populations in which the dominant white gene is segregating. The objective of this study was to investigate whether there is a gene that is responsible for deafness as well as for blue eyes and to establish a plausible mode of inheritance. For this purpose, data from an experimental colony with deaf cats were analyzed. The hearing status was determined by acoustically evoked brain stem responses (BAER). Complex segregation analyses were conducted to find out the most probable mode of inheritance using maximum likelihood procedures. The prevalence of deafness and partial hearing in the experimental colony was 67% and 29%, respectively. The results of the bivariate segregation analysis support the hypothesis of a pleiotropic major gene segregating for deafness and blue iris colour. The high heritability coefficients for both traits, 0.55 and 0.75 respectively, indicate that beside the major gene there is an important influence of polygenic effects.

EmsB, a tandem repeated multi-loci microsatellite, new tool to investigate the genetic diversity of Echinococcus multilocularis

In order to explore the genetic diversity within Echinococcus multilocularis (E. multilocularis), the cestode responsible for the alveolar echinococcosis (AE) in humans, a microsatellite, composed of (CA) and (GA) repeats and designated EmsB, was isolated and characterized in view of its nature and potential field application. PCR-amplification with specific primers exhibited a high degree of size polymorphism between E. multilocularis and Echinococcus granulosus sheep (G1) and camel (G6) strains. Fluorescent-PCR was subsequently performed on a panel of E. multilocularis isolates to assess intra-species polymorphism level. EmsB provided a multi-peak profile, characterized by tandemly repeated microsatellite sequences in the E. multilocularis genome. This "repetition of repeats" feature provided to EmsB a high discriminatory power in that eight clusters, supported by bootstrap p-values larger than 95%, could be defined among the tested E. multilocularis samples. We were able to differentiate not only the Alaskan from the European samples, but also to detect different European isolate clusters. In total, 25 genotypes were defined within 37 E. multilocularis samples. Despite its complexity, this tandem repeated multi-loci microsatellite possesses the three important features for a molecular marker, i.e. sensitivity, repetitiveness and discriminatory power. It will permit assessing the genetic polymorphism of E. multilocularis and to investigate its spatial distribution in detail.

A unification of models for meta-analysis of diagnostic accuracy studies

Studies of diagnostic accuracy require more sophisticated methods for their meta-analysis than studies of therapeutic interventions. A number of different, and apparently divergent, methods for meta-analysis of diagnostic studies have been proposed, including two alternative approaches that are statistically rigorous and allow for between-study variability: the hierarchical summary receiver operating characteristic (ROC) model (Rutter and Gatsonis, 2001) and bivariate random-effects meta-analysis (van Houwelingen and others, 1993), (van Houwelingen and others, 2002), (Reitsma and others, 2005). We show that these two models are very closely related, and define the circumstances in which they are identical. We discuss the different forms of summary model output suggested by the two approaches, including summary ROC curves, summary points, confidence regions, and prediction regions.

Health risk appraisal in older people 2: the implications for clinicians and commissioners of social isolation risk in older people

BACKGROUND: Social isolation is associated with poorer health, and is seen by the World Health Organisation (WHO) as one of the major issues facing the industrialised world. AIM: To explore the significance of social isolation in the older population for GPs and for service commissioners. DESIGN OF STUDY: Secondary analysis of baseline data from a randomised controlled trial of health risk appraisal. SETTING: A total of 2641 community-dwelling, non-disabled people aged 65 years and over in suburban London. METHOD: Demographic details, social network and risk for social isolation based on the 6-item Lubben Social Network Scale, measures of depressed mood, memory problems, numbers of chronic conditions, medication use, functional ability, self-reported use of medical services. RESULTS: More than 15% of the older age group were at risk of social isolation, and this risk increased with advancing age. In bivariate analyses risk of social isolation was associated with older age, education up to 16 years only, depressed mood and impaired memory, perceived fair or poor health, perceived difficulty with both basic and instrumental activities of daily living, diminishing functional ability, and fear of falling. Despite poorer health status, those at risk of social isolation did not appear to make greater use of medical services, nor were they at greater risk of hospital admission. Half of those who scored as at risk of social isolation lived with others. Multivariate analysis showed significant independent associations between risk of social isolation and depressed mood and living alone, and weak associations with male sex, impaired memory and perceived poor health. CONCLUSION: The risk of social isolation is elevated in older men, older persons who live alone, persons with mood or cognitive problems, but is not associated with greater use of services. These findings would not support population screening for individuals at risk of social isolation with a view to averting service use by timely intervention. Awareness of social isolation should trigger further assessment, and consideration of interventions to alleviate social isolation, treat depression or ameliorate cognitive impairment.