PubMed Health⌕ Search

Biomedical subjects

K Linnet

Publications and source records attributed to K Linnet.

18 recordsLinked to original sources
PubMed1991

Mean and variance rules are more powerful or selective than quality control rules based on individual values.

Quality control rules based on individual values are compared with mean and variance rules using theoretical computations and simulations. Simple (1(3)s) and combined individual value rules, e. g. a 1(3)s/2(2)s/4(1)s/6 means rule, are all less powerful for detection of shifts of location than a mean rule, given identical type I errors. The mean rule is also more robust towards non-normality of data distributions. In most cases, the variance rule has more power towards increased scatter than individual value rules, and it always has the highest selectivity. Thus, the simple computations that are required for derivation of the mean and variance result in increased power or selectivity. In particular, in the computerization of quality control, the traditional mean and variance rules are preferable to more or less complicated "multi-rules" proposed for computerized quality control.

Analysis of Variance↗
PubMed1991

HPLC with enzymatic detection as a candidate reference method for serum creatinine.

We present a candidate Reference Method for determining the concentration of serum creatinine. The method is based on HPLC combined with enzymatic determination. Creatinine plus 14[C]creatinine is extracted by cation-exchange chromatography, subjected to reversed-phase HPLC, and finally quantified enzymatically. Enzymatic measurement ensures no interference from co-eluting compounds, which has been a problem for some reported HPLC assays relying on ultraviolet detection. The average corrected recovery was 100.1% (SEM = 1.1%; n = 15). The accuracy was verified by assaying five sera with target values determined by isotope dilution mass spectrometry. The total interassay CV was less than or equal to 2.5%. We applied the method to study the specificity of HPLC-ultraviolet detection, using 72 plasma samples from hospitalized patients; no interference was noted. Thus, HPLC-ultraviolet detection appears to be specific, provided that sample cleanup is based on cation-exchange chromatography. Our diode-array detector studies of peak homogeneity supported this conclusion. Still, combined HPLC-enzymatic determination ensures even greater accuracy, ranking close to that by isotope dilution mass spectrometry.

Chromatography, High Pressure Liquid↗
PubMed1990

Estimation of the linear relationship between the measurements of two methods with proportional errors.

The linear relationship between the measurements of two methods is estimated on the basis of a weighted errors-in-variables regression model that takes into account a proportional relationship between standard deviations of error distributions and true variable levels. Weights are estimated by an interative procedure. As shown by simulations, the regression procedure yields practically unbiased slope estimates in realistic situations. Standard errors of slope and location difference estimations are derived by the jackknife principle. For illustration, the linear relationship is estimated between the measurements of two albumin methods with proportional errors.

Chemistry, Clinical↗
PubMed1989

Assessing diagnostic tests by a strictly proper scoring rule.

Evaluation of univariate quantitative diagnostic tests by strictly proper scoring rules is considered as an alternative to the traditional error rate measures. In principle, the posterior probability of disease as a function of the test value is estimated from training observations, and subsequently the score is assessed on a set of test samples. The same subjects may serve as training and test samples when the bootstrap procedure is applied for estimation of standard errors and correction of bias. The method is demonstrated using serum bile acids and bilirubin in patients with liver disease. The power for comparison of scores from two tests is compared with that from error rate measures for some typical situations.

Bile Acids and Salts↗
PubMed1989

The between-run component of variation in internal quality control.

Design of control charts for the mean, the within-run component of variance, and the ratio of between-run to within-run components of variance is outlined. The between-run component of variation is the main source of imprecision for analytes determined by an enzymo- or radioimmunoassay principle; accordingly, explicit control of this component is especially relevant for these types of analytes. Power curves for typical situations are presented. I also show that a between-run component of variation puts an upper limit on the achievable power towards systematic errors. Therefore, when the between-run component of variation exceeds the within-run component, use of no more than about four controls per run is reasonable at a given concentration.

Analysis of Variance↗
PubMed1989

Choosing quality-control systems to detect maximum clinically allowable analytical errors.

Critical systematic and random analytical errors for 17 common clinical chemical components were estimated from published values for analytical imprecision, biological variation, and "medically important changes." Appropriate quality-control systems for these analytes are discussed on the basis of power considerations. The simple rule 1(3)s, with one control per run, is minimally sufficient for the analytes (about one quarter of those considered here) for which the magnitude of critical error is at least 3 analytical standard deviations. The more powerful rule 1(2)s, with one control per run, is the minimal requirement for analytes for which critical errors are about 2 analytical standard deviations; these are about half the remaining analytes. Greater power values are achieved by using multiple rules based on several controls per run. In general, this study does not support the view put forward by some authors that the quality-control rules in use today are too restrictive.

Chemistry, Clinical↗
PubMed1988

On the sensitivity of linear discriminant analysis to sampling variation and analytical errors.

The influence of analytical inaccuracy and imprecision on the linear discriminant function is considered. Analytical shifts occurring between the analysis of samples from each of two groups give spuriously low error rates if the function is evaluated on the training set, notably at high dimensions. Inaccuracy arising after the establishment of a discriminant function may change considerably the individual group error rates whereas the overall error rate is moderately affected. Imprecision decreases the group separation by an amount comparable to that in the univariate situation. In conclusion, evaluation of the error rates of a discriminant function on an independent test set is important to obtain realistic estimates of the performance and is preferable to using unbiased statistical methods or the split-sample principle based solely upon the training set.

Biometry↗
PubMed1988

A review on the methodology for assessing diagnostic tests.

Evaluation of diagnostic tests by the following principles are reviewed: error rates, scores based on posterior probabilities, and the excess loss considered in a decision theoretic context. Error rates or the complementary non-error rates, specificity and sensitivity, are simple measures which provide a rough indication of the discriminative value. In clinical practice, where a test serves as a decision support together with other information, conversion of test results to posterior probabilities is recommended. An aggregate score of these probabilities expresses the value of the test. Finally, in simple, well-defined cases--for example, screening situations, where the prevalence of disease and the relative consequences of false-positive and -negative classifications can be estimated--a Bayesian decision analysis is appropriate. The optimal discrimination limit is selected, and the total loss is minimized. The likelihood ratio LR(x) plays a central role in probability calculations and in the decision analysis. An example illustrates application of the procedures.

Algorithms↗
PubMed1987

Comparison of quantitative diagnostic tests: type I error, power, and sample size.

For a quantitative laboratory test the 0.975 fractile of the distribution of reference values is commonly used as a discrimination limit, and the sensitivity of the test is the proportion of diseased subjects with values exceeding this limit. A comparison of the estimates of sensitivity between two tests without taking into account the sampling variation of the discrimination limits can increase the type I error to about seven times the nominal value of 0.05. Correct statistical procedures are considered, and the power and required sample size are studied for Gaussian and log-Gaussian distributions of diagnostic test values. The results may be useful for the planning phase of studies to evaluate quantitative diagnostic tests.

Biometry↗
PubMed1987

Two-stage transformation systems for normalization of reference distributions evaluated.

In two-stage transformation systems for normalization of reference distributions, the asymmetry is first corrected, and any deviation of kurtosis is then adjusted. The simulation studies reported here show that these systems have previously been assessed too optimistically because the sample variation of the transformation parameters was neglected. Applying a goodness-of-fit test to transformed values shows that one should accept gaussianity only for p-values greater than 0.15 instead of those greater than 0.05. Further, the calculated 90% confidence intervals of reference limits should be expanded by 25%. When the correct level of significance is used, only real reference distributions that deviate moderately from the gaussian form are normalized. Calculation of confidence intervals demonstrated that 50 to 450 subjects are needed for a precise parametric estimation of the 95% reference interval. For the nonparametric approach, 125 to 700 reference subjects are necessary. The larger sample sizes are needed when distributions show pronounced skewness.

Biometry↗
PubMed1986

Assessing diagnostic tests once an optimal cutoff point has been selected.

The specificity and sensitivity of a quantitative diagnostic test depends on the chosen cutoff point. The common practice of selecting a cutoff point that maximizes the specificity plus the sensitivity, as judged from the observed test results, is studied here by simulation. Test performance is on average assessed too optimistically by this procedure--a phenomenon of importance when sample sizes are small. For example, the average positive bias is up to 15% of the test performance for sample sizes of 25. Furthermore, binomial calculated standard errors of specificity and sensitivity estimates are incorrect. A Monte Carlo statistical method--the "bootstrap procedure"--is applied to correct for bias and to estimate standard errors, including the standard error of the optimal cutoff point. Independent and paired comparisons of two diagnostic tests are also considered when optimal cutoff points have been selected. For this purpose, binomial statistical tests behave satisfactorily. Examples of power functions are presented.

Bile Acids and Salts↗
PubMed1985

Analysis of insulin receptors on heterogeneous eukaryotic cell populations with fluorochrome-conjugated insulin and fluorescence-activated cell sorter. Advantages and limitations to the 125I-labelled insulin methodology.

The diversity in insulin receptor expression within eukaryotic cell populations can be studied with fluorochrome conjugated reagents with high affinity to the insulin receptor in combination with flow cytometry. We studied the optimal conditions for application of fluorescein isothiocyanate (FITC) conjugated insulin in combination with the fluorescence activated cell sorter (FACS) to analyse insulin receptor expression, and also studied the feasibility of this method for identifying and isolating viable subsets with differences in insulin receptor expression within a cell population. Semisynthetic human insulin was conjugated to FITC, which resulted in at least four types of FITC-insulin molecules with different affinities to the insulin receptor. Each type of FITC-insulin was isolated by semipreparative reverse phase high pressure liquid chromatography. The preparation with a fluorescein/protein ratio of approximately 1.0 was found to have the highest affinity to the receptor, the highest biological activity (approximately 50% of native insulin), and similar antigenicity as native insulin. The optimal staining conditions with respect to pH, time of incubation, and cell number were determined, and were different in some aspects from labelling with 125I-insulin. The binding of FITC-insulin to cells was saturable and could be displaced with unlabelled insulin. The fluorescence signal could be converted to absolute numbers of fluorescein molecules by a calibration curve, and the absolute number of specifically bound FITC-insulin molecules calculated from a F/P ratio approximately 1.0. The FITC-insulin/FACS method permits estimation of the total number of insulin receptors (high plus low affinity), and the data obtained correlate well with the results from Scatchard plot of 125I-insulin binding data.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1985

Precision of sensitivity estimations in diagnostic test evaluations. Power functions for comparisons of sensitivities of two tests.

The precision of estimates of the sensitivity of diagnostic tests is evaluated. "Sensitivity" is defined as the fraction of diseased subjects with test values exceeding the 0.975-fractile of the distribution of control values. An estimate of the sensitivity is subject to sample variation because of variation of both control observations and patient observations. If gaussian distributions are assumed, the 0.95-confidence interval for a sensitivity estimate is up to +/- 0.15 for a sample of 100 controls and 100 patients. For the same sample size, minimum differences of 0.08 to 0.32 of sensitivities of two tests are established as significant with a power of 0.90. For some published diagnostic test evaluations the median sample sizes for controls and patients were 63 and 33, respectively. I show that, to obtain a reasonable precision of sensitivity estimates and a reasonable power when two tests are being compared, the number of samples should in general be considerably larger.

Clinical Laboratory Techniques↗