Survival ratio plots with permutation envelopes in survival data problems.

A new method is proposed, based on a simple ratio plot, which is useful for identifying time intervals or regions in which survival prospects differ between two distinct populations or treatments. In order to investigate whether any observed difference or trend is due to sampling variation or is due to a possible real effect, resampling techniques are used to generate permutation envelopes as reference bands. The method is applicable to the comparison of survival in two populations for both independent and paired survival problems and also to the assessment of the symmetry of a bivariate survival function.

The influence of season and leaf age on concentrations of radiocaesium (137Cs), stable caesium (133Cs) and potassium in Agrostis capillaris.

The transfer of radioactive caesium from soils to plants has been well researched. In contrast there is limited knowledge on natural stable 133Cs and its potential role as a predictor for radiocaesium behaviour. In a pot experiment with Agrostis capillaris close correlations were found between plant 137Cs and plant 133Cs concentrations (R2 90-96%). Season and leaf age had significant effects with concentrations increasing 10-30-fold between June and December. Simultaneously the plant concentrations of K, the nutrient analogue of Cs, decreased to around one third. In the soil the exchangeable fractions of K and 137Cs declined. No clear relationships were found between 137+133Cs in the plant and exchangeable K in the soil. However, at the end of the experiment the K content of the above-ground biomass was higher than the exchangeable pool in the soil, suggesting that depletion of soil K could be a key factor in the observed increase of plant 137+133Cs over time.

Climbing-specific finger endurance: a comparative study of intermediate rock climbers, rowers and aerobically trained individuals.

The aim of this study was to compare the climbing-specific finger endurance of climbers, rowers and aerobically leg trained athletes. Twenty-seven males aged 21.2 +/- 2.2 years (mean +/- s) volunteered for the study. The participants were intermediate rock climbers (n = 9), rowers (n = 9) and leg trained athletes (n = 9). Maximal voluntary contraction (MVC) was determined on climbing-specific finger apparatus. Endurance isometric exercise was performed at 40% MVC in three tests performed in a random order: (1) sustained exercise; (2) 6 s exercise, 4 s rest; and (3) 18 s exercise, 12 s rest. Pre- and post-exercise blood pressure and blood lactate concentration, together with post-exercise pain perception, were measured. The climbers had a significantly greater MVC (383 +/- 35.6 N) than the rowers (321 +/- 49.5 N, P = 0.007) and aerobically leg trained athletes (288 +/- 60.6 N, P = 0.001). There were no significant differences between the groups in terms of endurance times for any of the tests. In the test with 18 s exercise and 12 s rest, the climbers showed a significantly higher increase in blood lactate concentration, on average, than the rowers by 0.01-0.89 mmol x l(-1) (P = 0.006); there were no significant differences, on average, in the comparisons of climbers and the leg trained athletes and rowers and the leg trained athletes. There were no significant differences in the average changes in blood pressure from rest to post-exercise between any of the groups. Although the climbers had greater MVC on average than the other two groups, there were no significant differences in average endurance times amongthe groups. These findings suggest that training for rock climbing and participation in rock climbing may result in some specific adaptations. However, we acknowledge that this study is descriptive and there is the possibility that differences between groups could be attributed to self-selection.

Sea level and acute responses to hypoxia: do they predict physiological responses and acute mountain sickness at altitude?

OBJECTIVES: To compare a range of physiological responses to acute sea level hypoxia at simulated altitudes with the same physiological responses and acute mountain sickness (AMS) scores measured at altitude (similar to the simulated altitudes) during a 17 day trek in the Himalayas. METHODS: Twenty men and women aged 18-54 years took part in the study. End tidal CO(2) tension (PETCO(2)) and saturated oxygen (SaO(2)) were measured using a capnograph. Observations made at sea level and some simulated altitudes were compared with those in the Himalayas. Pairwise correlations were used to examine the correlation between variables and regression, with 95% prediction intervals providing information on how well one variable could be predicted from another for a given subject. RESULTS: There was only a significant correlation for a few comparisons. The 95% prediction intervals for individual SaO(2) values at a range of simulated altitudes were fairly wide going from +/- 4% to +/- 5%. All of the correlations between laboratory and Himalayan PETCO(2) values were not statistically significant. There was a significant correlation for the relation between SaO(2)and PETCO(2) at sea level for the laboratory data (r=-0.51; p=0.05). For the Himalayan data, there were significant correlations at Namche Bazaar (3450 m; day 3) (r=-0.56; p=0.01) and Dingboche (4300 m; day 6) (r=-0.48; p=0.03). The correlation between SaO(2) and PETCO(2) and AMS was generally poor. CONCLUSIONS: The results indicate that limited information can be gained on a subject's response to altitude by assessing physiological variables at sea level and a range of simulated altitudes before the subject carries out a trek at altitude.

Discrimination of the notifiable pathogen Gyrodactylus salaris from G. thymalli (Monogenea) using statistical classifiers applied to morphometric data.

The identification and discrimination of 2 closely related and morphologically similar species of Gyrodactylus, G. salaris and G. thymalli, were assessed using the statistical classification methodologies Linear Discriminant Analysis (LDA) and k-Nearest Neighbours (KNN). These statistical methods were applied to morphometric measurements made on the gyrodactylid attachment hooks. The mean estimated classification percentages of correctly identifying each species were 98.1% (LDA) and 97.9% (KNN) for G. salaris and 99.9% (LDA) and 73.2% (KNN) for G. thymalli. The analysis was expanded to include another 2 closely related species and the new classification efficiencies were 94.6% (LDA) and 98.% (KNN) for G. salaris; 98.2% (LDA) and 72.6% (KNN) for G. thymalli; 86.7% (LDA) and 91.8% (KNN) for G. derjavini; and 76.5% (LDA) and 77.7% (KNN) for G. truttae. The higher correct classification scores of G. salaris and G. thymalli by the LDA classifier in the 2-species analysis over the 4-species analysis suggested the development of a 2-stage classifier. The mean estimated correct classification scores were 99.97% (LDA) and 99.99% (KNN) for the G. salaris-G. thymalli pairing and 99.4% (LDA) and 99.92% (KNN) for the G. derjavini-G. truttae pairing. Assessment of the 2-stage classifier using only marginal hook data was very good with classification efficiencies of 100% (LDA) and 99.6% (KNN) for the G. salaris G. thymalli pairing and 97.2% (LDA) and 99.2% (KNN) for the G. derjavini-G. truttae pairing. Paired species were then discriminated individually in the second stage of the classifier using data from the full set of hooks. These analyses demonstrate that using the methods of LDA and KNN statistical classification, the discrimination of closely related and pathogenic species of Gyrodactylus may be achieved using data derived from light microscope studies.

The use of statistical classifiers for the discrimination of species of the genus Gyrodactylus (Monogenea) parasitizing salmonids.

This study applies flexible statistical methods to morphometric measurements obtained via light and scanning electron microscopy (SEM) to discriminate closely related species of Gyrodactylus parasitic on salmonids. For the first analysis, morphometric measurements taken from the opisthaptoral hooks and bars of 5 species of gyrodactylid were derived from images obtained by SEM and used to assess the prediction performance of 4 statistical methods (nearest neighbours; feed-forward neural network; projection pursuit regression and linear discriminant analysis). The performance of 2 methods, nearest neighbours and a feed-forward neural network provided perfect discrimination of G. salaris from 4 other species of Gyrodactylus when using measurements taken from only a single structure, the marginal hook. Data derived from images using light microscopy taken from the full complement of opisthaptoral hooks and bars were also tested and nearest neighbours and linear discriminant analysis gave perfect discrimination of G. salaris from G. derjavini Mikailov, 1975 and G. truttae Gläser, 1974. The nearest neighbours method had the least misclassifications and was therefore assessed further for the analysis of individual hooks. Five morphometric parameters from the marginal hook subset (total length, shaft length, sickle length, sickle proximal width and sickle distal width) gave near perfect discrimination of G. salaris. For perfect discrimination therefore, larger numbers of parameters are required at the light level than at the SEM level.

Towards an automated system for the identification of notifiable pathogens: using as an example.

Simple and rapid identification of pathogen species is crucial to the control of many diseases. Here, James Kay, Andrew Shinn and Christina Sommerville demonstrate that statistical classifiers discriminate a notifiable pathogen Gyrodactylus salaris Malmberg, 1957, a lethal ectoparasite of Atlantic salmon, Salmo salar L., from its benign close relatives.

Management of chronic airflow obstruction: differences in practice between respiratory and general physicians.

An audit of inpatient care of diseases characterized by chronic airflow obstruction namely chronic bronchitis, emphysema and chronic obstructive airways disease (ICD Code Nos. 490-2 & 496) was performed and the practice of respiratory and general physicians compared. One hundred cases were sampled at random from 279 cases admitted to hospitals serving the West of Glasgow in 1988. Fifty cases were selected from those admitted under the care of respiratory physicians and 50 from those under general physicians; 89 were suitable for analysis. The main outcome measurements consisted of the use of routine respiratory investigations, comparison of the use of standard therapies during the admission and at discharge, length of stay, inpatient deaths, follow up and readmission rates. The groups were similar in age, smoking history, gender and there was no significant difference in admission arterial blood gas values. The pulse rate on admission was higher in the general group (102 beats per min) in comparison to the respiratory group (91 beats per min) (P < 0.004). A similar use of chest radiograph and arterial blood gas analysis was noted between the groups. Ninety-six per cent of respiratory patients had either spirometry or peak expiratory flow measured compared to 62% in the general group (P = 0.0001). No significant differences were noted in the use of antibiotics, bronchodilators, corticosteroids, oxygen or respiratory stimulants. The mean length of stay was similar. Two patients (4%) in the respiratory group compared with seven (18%) in the general group died during the admission (P = 0.01); there were no further early deaths at 1 month from discharge.(ABSTRACT TRUNCATED AT 250 WORDS)

A comparison of treatment with metformin and gliclazide in patients with non-insulin-dependent diabetes.

Twenty-seven obese non-insulin-dependent diabetic patients, treated with dietary carbohydrate restriction and metformin, were recruited from the diabetic outpatient clinic and entered into an open crossover study with gliclazide. Twenty-one patients completed the study. During three months observation on metformin, the mean weight of the group fell by 1.0 kg with 14 patients losing a mean of 1.8 kg with 14 patients losing a mean of 1.8 kg, 3 remaining unchanged and 4 gaining a mean weight of 1.1 kg. Over the subsequent three months on gliclazide, the mean weight of the group rose by 1.4 kg with 16 patients gaining a mean of 2.2 kg, two remaining unchanged and 3 losing a mean of 2.0 kg. In addition, 10 patients were heavier after gliclazide than at the time of recruitment (mean 2.6 kg), 3 were unchanged and 8 had lost weight since commencing the trial (mean 2.1 kg), mostly due to greater loss on metformin than gain on gliclazide. Glycaemic control did not improve significantly during the observed period on metformin but lower concentrations of fasting glucose and total glycosylated haemoglobin were achieved with gliclazide. Mean plasma insulin concentration was significantly higher and mean serum lactate was significantly lower during treatment with gliclazide. In conclusion, gliclazide does not support weight loss in obese non-insulin-dependent diabetic patients to the same extent as metformin but the difference between the two drugs is small. Gliclazide is a suitable oral hypoglycaemic agent for use in the obese diabetic who cannot be controlled by diet alone.

Importance of western blot analysis in predicting infectivity of anti-HTLV-III/LAV positive blood.

Stored donor and recipient sera from prospective studies of post-transfusion hepatitis were analysed for the presence of human T-cell lymphotropic virus type-III/lymphadenopathy associated virus (HTLV-III/LAV) antibodies as determined by enzyme-linked immunosorbent assays (ELISA). Of 3961 donor samples given to 461 patients, only 2 (0.05%) contained specific HTLV-III/LAV antibodies as determined by an avidin-biotin-enhanced western blot tech nique. Anti-HTLV-III/LAV was measured before and 3 and 6 months after transfusion in 295 recipients of anti-HTLV-III-negative blood, 7 recipients of ELISA-positive blood which was western blot negative, and 2 recipients of ELISA-positive blood confirmed as specific by western blot. Only the last 2 recipients became infected with HTLV-III/LAV, as assessed by antibody seroconversion (p less than 0.0001). Serocon version occurred early (6 and 8 weeks after transfusion) and was characterised first by antibody to p24 and later by antibody to p41. AIDS has not developed in either patient, but one has a T4/T8 ratio of 0.4 and impaired mitogen responses; the second patient has no evidence of immune dysfunction 4 years after exposure. This study confirms that HTLV-III/LAV infection can be transmitted by blood transfusion and supports the advisability of anti-HTLV-III/LAV testing of all blood donors. It also confirms the validity of western blot testing for HTLV-III/LAV specificity and suggests that ELISA-positive, western-blot-negative blood may not be infectious.