Diagnostic tests accuracy

Irwig L, Macaskill P, Glasziou P, Fahey M, Meta-analytic methods for diagnostic test accuracy. J Clin Epidemiol 1995 48 119-30. 

A new tumor marker is evaluated using the same criteria used for many diagnostic tests (i.e., sensitivity, specificity, and accuracy). The diagnostic sensitivity and specificity are best represented by a receiver operating characteristic (ROC) curve. The ROC curve is constructed with the true-positive rate versus false-positive rate at various decision levels. As a test improves in its diagnostic performance, it shifts upward and to the left as the true-positive rate increases and the false-positive rate decreases. 

By best research evidence Sackett means clinically relevant research, often from the basic sciences of medicine, but especially from patient centred clinical research into the accuracy and precision of diagnostic tests (including the clinical examination), the power of prognostic markers, and the efficacy and safety of therapeutic,... 

Cochrane methodology reviews As well as Cochrane Reviews of the effects of healthcare interventions, there are also Cochrane methodology reviews of the ways in which health care can be evaluated and, from 2006, there will be Cochrane Reviews of the accuracy of diagnostic tests. 

Chest x-ray, computed tomography (CT) scan, and positron emission tomography (PET) scan are the most valuable diagnostic tests. Integrated CT-PET technology appears to improve diagnostic accuracy in staging NSCLC over CT or PET alone. 

Among the steps in conducting a systematic review of a diagnostic test (Box 13-1), the most important is the identification of the chnical question for which the test result is required to give an answer, and thus formulation of the question that forms the basis of the review. Two types of questions can be addressed in a systematic review in diagnostic medicine one type is related to the diagnostic accuracy of a test, and the other to the clinical value (to patients or to others) of using the test. The questions that arise are similar in structure but require different approaches. 

Until recently, methodologists interested in systematic reviews have focused on studies of the effects of interventions, especially drugs, on patient outcomes. That work is generally applicable to systematic reviews of diagnostic tests that start with a question of the second type above. The opportunities to use the techniques, however, are limited as it is unusual to find more than one study on any combination of a test and an outcome. We therefore focus on systematic reviews of the diagnostic accuracy of tests and express the hope that the quality of primary studies will continue to improve so that many such systematic reviews will become feasible. 

The challenge in diagnosing CHR is that many of the presenting symptoms are nonspecific. Yet, studies have shown that there were dramatic differences in BNP of patients with dyspnea caused by cardiac failure (mean of 1076 ng/L) versus BNP in patients with dyspnea caused by noncardiac etiologies (mean of 86 ng/L).ROC curves have shown that BNP had better overall accuracy than the clinician s judgment (AUC 0.97 versus 0.88), with similar findings observed for NT-proBNP.It is important to remember that BNP is not a stand-alone diagnostic test it must be used and... 

Other critics state that EBM considers randomized, controlled trials (RCTs) as the only evidence to be used in clinical decision making. AcmaUy, EBM seeks the best existing evidence from basic science to clinical research with which to inform clinical decision. For example, a decision about the accuracy of a diagnostic test is best informed by evidence from a cross-sectional study, not an RCT. A cohort study, not an RCT, best answers a question about prognosis. However, in selecting a treatment, the randomized trial is the best study design to provide the most accurate estimate of treatment efficacy and safety. 

With the increased availability of a variety of home diagnostic tests, the sale of home pregnancy, ovulation, and fertility tests alone was a 330 million dollar business in 2005. Home pregnancy tests have been utilized since 1976, with approximately 26 devices currently available in the United States. These tests claim that they can be utilized on the first day of a missed menstrual period (or, for oertain devioes, even earlier) with more than 99% diagnostic accuracy for pregnancy. Despite these claims, accurate results are obtained only a fraction of the time (Table 46.1) (1). 

As we have tried to point in this chapter, development of an ultrasensitive measurement device is only the first step in realizing broad based use of multiple-protein detection in cancer diagnostics. As illustrated in Figs. 1.5, 1.6, 1.11 and 1.13, an immediate concern is accuracy in the sample medium to be used in the diagnostic test. That is, the new device should be tested with real samples against accurate alternative methods such as ELISA, and good correlations obtained to ensure accuracy. 

Ostrowski JA (2006) Accuracy of 3 diagnostic tests for anterior cruciate ligament tears. J Athl Train 41 120-121... 

Scholten RJ, Opstelten W, van der Plas CG, Bijl D, Deville WL, Bouter LM (2003) Accuracy of physical diagnostic tests for assessing ruptures of the anterior cruciate ligament a meta-aniysis. J Fam Pract 52 689-694... 

Other issues With software detection, it is not possible to localize the error sources. So, diagnostic test programs are frequently used identify the module concerned. Almost all DCS have some form of diagnostic routines to pinpoint the faults as best as possible. The accuracy of such pinpointing may vary from system to system. In a fault tolerant system, the system itself initiates these tests and interprets their results, as opposed to the outside insertion of test programs by operators in other systems [4]. 

However, because there is no current gold standard for the diagnosis of H. pylori, the choice of diagnostic test depends more on local resources, experience, and cost-effectiveness than small differences in sensitivity and specificity. Probably, serology will be used to establish infection and breath test for efficacy of eradication. However, the accuracy of many of the current serology tests could be improved. False positives lead to needless expense, false negatives to inappropriate peace of mind on the part of either patient or physician. 

The basic aim of medical research is to improve clinical practice, and this should be evidence based, if possible. Clinically relevant research evidence may relate to basic medical science, but it especially relates to patient-centered cUnical research, e.g., to the study of the accuracy and precision of diagnostic tests, the power of prognostic markers, and the efficacy and safety of therapeutic, rehabilitative, and preventive regimens. New evidence from clinical research at the same time invalidates previously accepted diagnostic tests and treatments and replaces them with new ones that are more powerful, more accurate, more efficacious, and safer [1]. Ethical decision making is based on the Declaration of Helsinki,... 

Marquette CH, Copin MC, Wallet F, et al. Diagnostic tests for pneumonia in ventilated patients prospective evaluation of diagnostic accuracy using histology as a diagnostic gold standard. Am J Respir Crit Care Med 1995 151 1878-1888. 

The main development in medical diagnostic reagents since the 1960s has been the steady growth of dry (soHd-phase) chemistry systems. Dry chemistry systems have made substantial gains over wet clinical analysis in the number of tests performed in hospitals, laboratories, and homes because of ease, rehabiUty, and accuracy. 

The diagnostic usefulness should be established by appropriate retrospective or prospective studies to see if the test in fact improves the diagnostic accuracy. For inherited enzyme diseases, the measurement of the enzyme activity can be diagnostically specific and highly useful. For otherwise obvious diseases such as metastatic prostatic carcinoma, enzyme tests may be of little value (24). 

---