Showing papers on "False positive paradox published in 1990"

Development of a Record Linkage Protocol for Use in the Dutch Cancer Registry for Epidemiological Research

Abstract: A method has been developed to determine the optimal linkage key for record linkage between the cancer registry and a large-scale prospective cohort study in the Netherlands. The proposed linkage procedure is a two-stage process in which the initial computerized linkage using a particular linkage key is followed by visual inspection with additional information to separate the computer matches into true and false positives. In the determination of the optimal key, both informativeness and susceptibility to error of personal identifiers were taken into account. The performance of the various keys in the linkage was expressed in terms of sensitivity and predictive value of a reported computer match. The key, consisting of date of birth, first four characters of the family name and gender was the optimal choice, with a sensitivity of 98% and an initial predictive value of a computer match of 98%. When additional information on migration, place of birth and first initial was collected in the second stage, it was possible to eliminate the false positives from the reported computer matches without loss of true positives. Thus, the sensitivity remained constant whereas the secondary predictive value of accepted matches was maximized.

Detecting disease clusters: the importance of statistical power

Abstract: A variety of methods and models have been proposed for the statistical analysis of disease excesses, yet rarely are these methods compared with respect to their ability to detect possible clusters. Evaluation of statistical power is one approach for comparing different methods. In this paper, the authors study the probability that a test will reject the null hypothesis, given that the null hypothesis is indeed false. They present a discussion of some considerations involved in power studies of cluster methods and review two methods for detecting space-time clusters of disease, one based on cell occupancy models and the other based on interevent distance comparisons. The authors compare these approaches with respect to: 1) the sensitivity to detect disease excesses (false negatives); 2) the likelihood of detecting clusters that do not exist (false positives); and 3) the structure of a cluster in a given investigation (the alternative hypothesis). The methods chosen, which are two of the most commonly used, are specific to different hypotheses. They both show low power for the small number of cases which are typical of citizen reports to health departments.

Apparatus and method for determining the ability of an individual to perform a task

Abstract: An apparatus and method for testing individuals to determine their capability of conducting predetermined tasks. A base line is generated which is unique to each individual to whom the test is to be presented. The individual's current response to the battery of tests is compared by using an algorithm with the individual's base line to determine impairment of the individual. The algorithm is statistically designed to pass individuals relative to their base line performance level while eliminating all false negatives. When an individual tests positive, the test is repeated one or more times and the results compared using an algorithm which is statistically designed to pass individuals relative to their individual base line performance levels while eliminating all false positives. Such is accomplished by providing a source of images randomly and non-repetitively to a screen for recognition and response by the individual being tested. The responses are recorded in a compiler means and then compared to the previous results of the same individual which are stored in a data base and then tabulated to provide an output signal indicative of the individual's present ability to perform the task in question.

Computer-aided diagnosis of breast aspirates via expert systems.

Abstract: Two computer-driven expert systems trained to correctly diagnose 369 fine needle aspirates of the breast on the basis of nine cytologic descriptive parameters were tested on 70 newly obtained aspirates (57 benign and 13 malignant). The system generated by multisurface pattern separation misclassified one malignant test sample (i.e., one false negative) while the system generated by a connectionist algorithm (neural network) misclassified two benign test samples (i.e., two false positives). A decision tree misclassified three of the benign test samples (i.e., three false positives). These expert systems aid in the cytologic diagnosis of breast aspirates and can serve as models for other applications.

Value-of-information analysis of testing strategies: estimating the effect of uncertainty about the proportion of chemicals that are true human carcinogens.

Abstract: Choosing a cost-effective strategy for classifying chemicals as human carcinogens and non-carcinogens depends upon the costs of false positives (carcinogens erroneously treated as non-carcinogenic) and false negatives (non-carcinogens erroneously treated as carcinogenic); upon the accuracy (sensitivity and specificity) of the classification strategy; and upon the underlying proportion of carcinogens in the population of chemicals to be classified. If these values are known, value-of-information analyses can indicate the most cost-effective among three strategies: classify as carcinogenic without testing, classify as non-carcinogenic without testing, or choose the most cost-effective test and classify on the basis of the test result. When some or all of the values are uncertain, the analysis becomes more complex, but still helps to guide decisions among the three classification strategies.

Drug testing of physicians: the danger of false positives.

Abstract: To the Editor.— Dr Orentlicher 1 of the American Medical Association's Office of the General Counsel appropriately considers the issue of false-negative results in his discussion of drug testing of physicians. He fails, however, to mention the equally important issue of false-positive results, apparently assuming that "confirmatory tests (mass spectrometry and chromatography)" are incapable of error. It is true that these techniques have admirably low rates of false positivity, less than 0.01%. 2 But it is axiomatic in laboratory medicine that tests are inevitably imperfect; neither absolute sensitivity nor absolute specificity can be achieved. It should also be obvious that no amount of chain-of-custody documentation can entirely prevent the occasional contamination or exchanging of samples. I have never used illicit drugs or misused legal ones. The probability of a true-positive test of my urine is exactly zero. The probability of a false-positive test, no matter how small mathematically, is finite

Contribution of motor performance tests to the early diagnosis of Parkinson’s disease

Abstract: With the help of multivariate statistical methods, we analysed the results of a motor performance test series consisting of simple subtests. We were able to differentiate almost completely between parkinsonian patients (most of whom showed only slight or very slight clinical symptoms) and age-matched controls: we found only 2% false negative diagnoses and no false positives. Thus, impairment of fine motor skills is very frequent in the early stages of the disease and standardized motor performance tests can contribute to its detection. Since we established that a deterioration of the test results is highly age-correlated for these patients, we expect that follow-up testing will further contribute to improve selectivity.

Evaluation of simple tests for detection of HIV antibodies: analysis of interobserver variation in Tanzania.

Abstract: 200 sera were tested for HIV antibodies with different tests in Tanzania. The results were interpreted by 5 different observers with different laboratory experience. There was considerable variation between observers and between testing methods. HIV-Chek was easiest to perform with little interobserver variation, but a few probably false negative readings were seen. Serodia gave many false positives. Western Blots gave more diverging results due to indeterminate sera and lack of training. HIV-Chek seems best when time, facilities, and training are limited and if combined with Serodia false negative results can be excluded.

[The application of medical decision making to the mass screening: the basic principles on ROC analysis].

Abstract: The basic principles on receiver-operating characteristic (ROC) analysis were discussed. ROC curves showed the discriminative ability of a test by the position of the full curve in a graph plotting the relation between the true positive rate (TPR) and the false positive rate (FPR) over a wide range of cut-off points of a test. The increase in the area under the ROC curve, or the shift of the curve upward and to the left in the diagram means that the test has better discriminative ability. A manual was given to conduct the ROC analysis with special reference to calculation of TPR, FPR, and the area under the ROC curve. Also discussed was the method to decide the best or optimal cut-off point as the positivity criterion of a test, based on the ROC analysis. Attention was paid to balance the risk of false negatives and false positives. We made an equation to decide the best cut-off point, which showed us the variables to be considered in the analysis of cut-off problems: the prevalence of disease and the outcome associated with each state classified by the test, i.e., true positive, false positive, true negative, and false negative.

[Comparison of three methods for screening aflatoxin-producing strains].

Abstract: By using rice medium microcolumn method, aflatoxin-producing ability medium dish method (APA method), A and M medium thin-layer chromato scanner method (AM method), seventy-six strains of Aspergillus flavus from the foodstuffs in Zhejiang province were screened for aflatoxin-producing strains, and which were confirmed with thin-layer chromatography method. Using the above three methods, aflatoxin-producing strains rates were detected respectively in 11.8%, 9.2% and 18.4%. Twenty aflatoxigenic strains were determined positive by all three methods (26.3%). All three methods showed false negatives. The AM method and APA method both showed false positives. The causes of false negatives and positives were discussed and the accord-rate of AM method with other methods was the highest as compared with that of microcolumn or APA method. False negative rate was the highest with APA method and next in order was that of microcolumn method. False positives appeared in both APA method and AM method. The causes of false negative and positive were discussed. It was suggested that AM method be safe, simple, and reliable method screening for aflatoxigenic strains be it improved in some aspects.

Presentation of false-negative test results.

Abstract: Sir .—Taubman et al 1 in the January 1989 issue of AJDC discuss an extremely important issue—use of a rapid latex agglutination test for the detection of group A β-hemolytic streptococcal pharyngitis in an office setting. Their study design demonstrates the importance of comparing a new test with "the gold standard" (ie, a reference laboratory culture) as well as with the customary test (ie, throat cultures incubated and interpreted in the pediatric office). The authors introduce a term, " percent of false-negative results (% FN*)," that they define as follows: where TN indicates true negatives; TP, true positives; and FP, false positives. In a situation like this, the risk of undetected group A β-hemolytic streptococcal pharyngitis (ie, the false-negative tests) is usually communicated by the negative predictive value (PV neg ) of the test, an established and frequently used term in clinical epidemiology. If the authors wish to refer to the proportion of

Robust voice command interface design through the use of error templates

Abstract: A scheme for enhancing the robustness of conventional M-ary pattern classifiers in noisy, cluttered, and distorting environments by the addition of error templates is presented. Independent error templates implement a nonparametric clutter model and are effective in reducing false positive errors in the presence of structured disturbances, whereas linked error templates reduce category errors (falsely labeling data from category i as category j). To test the effectiveness of independent and linked error templates and to determine any problems that they may create, digital simulations were done. A hardware prototype using error templates was built and used by a handicapped client with speech and motor disabilities. The simulations show that the addition of error templates significantly decreases false positives, with some improvement in the other categories. Results for the hardware prototype are encouraging. >

Observations on 'False Negative' and 'False Positive' Rates

Abstract: One of the more discouraging trends in the literature of drug testing is the rate at which the toxicology community seems willing to accept imprecision in the use of its technical terms of art. This practice never fails to evoke derision when detected in lay commentary, but I am increasingly struck by the convergence of professional and lay literature in this respect. In the drug testing literature, I am particularly dismayed at the case with which the terms \"false positive\" and \"false negative\" are bandied about. My most recent experience with this phenomenon is provided by the March 1990 issue of Forensic Urine Drug Testing, the newsletter provided by AACC to laboratories participating in the Forensic Urine Drug Testing Survey and accreditation program. In that publication, I find on page 4 a presentation of the false positive and false negative error rates of drug-testing laboratories, as revealed by the CAP/AACC Survey. The data shown in tabular array are, to say the least, rather alarming if substantiated. The prospect of false positive rates in excess of 1~ much less 6070, in high-volume workplace urine drug testing programs does not bear contemplation. A careful reading of the text that accompanies the table, however, reveals a very different problem-at least it does to a reader familiar with the history of the AACC survey. It turns out that much of the laboratories' diffieulty attaches to Survey specimens spiked at or below common analytical cutoffs or to specimens whose \"corrrect\" analysis required reporting rules different from those normally used for workplace drug testing. Given the rather elliptical nature of the Survey instructions to participant labs, itis not terribly surprising to me that a higher-than-usual number of labs got momentarily lost in the turns. I would suggest, however, that this situation has less to do with the laboratories' analytical proficiency in testing workplace urine specimens than it does with their ability to keep up with the twists and turns in the political confrontation between different accreditation agencies. In your own Journal, I have noted a three-part installment on \"Further Observations: False Negative EMIT Cannabinoids,\" \"False Negative Rate on EMIT Cannabinoids,\" and \"More on the False Negative Rate for EMIT Cannabinoids\" (1,2,3). Now, the EMIT cannabinoids assay is certainly in wide enough use to make its properties of great concern to laboratorians. It is important to all of us that information on technical experiences with the assays …