Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “diagnostic thresholds”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 343 records · Page 19Linked to original sources
PubMed2002

[Radioguided surgery for primary hyperparathyroidism: 100 cases].

One hundred patients with primary hyperparathyroidism underwent radioguided surgery within 90 to 180 minutes following IV injection of a diagnostic dose of MIBI and after parathyroid scanning with planar and oblique views. Exclusion criteria were thyroid pathology requiring surgery and suspicion of multiple endocrine neoplasia. A>20% step-up between the background noise and tissue uptake was the diagnostic threshold for parathyroid hyperfunctionning tissue, and this was observed in only 15% of cases overall all cured, but two. This method of detection is technically demanding and various angles of application of the probe can result in significant discrepancies of recorded uptake for the same spot. Our results demonstrate a physiological step-up between the ipsilateral unaffected upper and lower quadrants of the neck (range: - 34% to - 5%), seemingly because of the proximity of supra-aortic vessels (upper neck-upper mediastinal gradient: - 57% to - 21%). Therefore, significant ratios are meaningful only between either the symmetrical left and right controlateral quadrants, 8% of pre-operative scannings were non-contributory, and probe detection was contributory in 3/8 cases only. Benefit of the technique is limited in routine, but it can be helpful in redo cases if the offending gland is not located in close surroundings of tissues physiologically taking up the radio pharmaceutical (salivary glands, great vessels and heart).

Humans↗
PubMed2002

Prediction of myocutaneous adverse side effect due to intra-arterial chemotherapy by intra-arterial (99m)Tc-macroaggregated albumin administration in patients with bone and soft-tissue tumors.

UNLABELLED: In malignant bone and soft-tissue tumors, intra-arterial chemotherapy and limb-saving surgery have become popular. Myocutaneous inflammatory change and necrosis are the major local side effects of intra-arterial chemotherapy. (99m)Tc-macroaggregated albumin (MAA) imaging with intra-arterial tracer administration was performed to evaluate drug distribution, and the ability of (99m)Tc-MAA imaging to predict local side effects was assessed. METHODS: In 24 patients, 42 (99m)Tc-MAA images were obtained with tracer injection through an intra-arterial catheter that was inserted into the proximal portion of the tumor-feeding artery. Abnormal uptake other than by tumor was assessed visually and quantitatively. RESULTS: In visual analysis, abnormal (99m)Tc-MAA accumulation was observed in 21 of 42 images. In the first consecutive 13 of these 21 images, intra-arterial chemotherapy with cisplatin, doxorubicin, and caffeine was administered, and myocutaneous inflammation or necrosis in the area corresponding to the abnormal (99m)Tc-MAA uptake was observed in 11. In contrast, none of the 21 images without abnormal (99m)Tc-MAA uptake demonstrated any local adverse effect from intra-arterial chemotherapy. In the last consecutive 8 images with abnormal (99m)Tc-MAA uptake, intra-arterial chemotherapy was initiated with only cisplatin, and doxorubicin and caffeine administration was changed to the intravenous route. In all 8 of these images, no local adverse effects from chemotherapy were observed. Overall, the sensitivity, specificity, and accuracy of (99m)Tc-MAA imaging for the detection of myocutaneous damage were 100% (11/11), 91% (21/23), and 94% (32/34), respectively, and positive and negative predictive values were 85% (11/13) and 100% (21/21), respectively. In quantitative analysis, when the diagnostic threshold of the uptake ratio was set at 2.5, sensitivity, specificity, and accuracy for the detection of myocutaneous complications were 91% (10/11), 96% (22/23), and 94% (32/34), respectively, and positive and negative predictive values were 91% (10/11) and 96% (22/23), respectively. CONCLUSION: (99m)Tc-MAA imaging with intra-arterial infusion before intra-arterial chemotherapy for bone and soft-tissue tumors can facilitate prediction of local myocutaneous adverse effects due to chemotherapy.

Adolescent↗
PubMed2002

[Dynamic and flexible classification of transplant rejection].

Different classification systems of renal allograft rejection are in use. Although, significant progress has been made with the introduction of the widely recognized Banff classification scheme, major problems remain unsolved. Minimal diagnostic threshold levels for tubulo-interstitial rejection are lacking. "Hyperacute, acute and chronic rejection or allograft nephropathy" are ill defined entities and thereby hamper rather than promote the communication between clinicians and pathologists and prevent a better understanding of pathophysiological events. Here, we propose a classification system of renal allograft rejection which is primarily based on well characterized morphological changes. All anatomical compartments of the renal cortex i.e., the tubulo-interstitial compartment, arteries and glomeruli, can be affected by rejection individually or in combination. In the subgroups only morphological changes known to be induced by rejection are considered. Such a classification system has, in our opinion, several major advantages. Every group and subgroup can easily be amended by "qualifiers" providing additional immunohistochemical and molecular-biological information as they are known today or become available in the future. As knowledge expands, qualifiers may be added or removed without fundamentally changing the entire classification scheme--making the system flexible and dynamic.

Graft Rejection↗
PubMed2004

Contribution of bone mineral density and bone turnover markers to the estimation of risk of osteoporotic fracture in postmenopausal women.

In osteoporosis, the main cause for concern is the increase in the risk of fractures. The level of bone mineral density (BMD) measured by various techniques has been shown to be a strong predictor of fracture risk in postmenopausal women. However, half of patients with incident fractures have BMD value above the diagnostic threshold of osteoporosis defined as a T-score of -2.5 SD or more below the average value of young healthy women. Clearly there is a need for improvement in the identification of patients at risk for fracture. Several prospective studies have shown that an increased bone resorption evaluated by specific biochemical markers was associated with increased risk of the hip, spine and non-vertebral fractures independently of BMD. The use of bone markers in individual patients may be appropriate in some situations, especially in women who are not detected at risk by BMD measurements. For example, in the OFELY study including 668 postmenopausal women followed prospectively over 9 years, we found that among the 115 incident fractures, 54 (47%) actually occurred in non-osteoporotic women. Among these women, the combination of bone markers and history of previous fracture was highly predictive of fracture risk. Thus, bone markers may be used in the assessment of fracture risk in selected cases in which BMD and clinical risk factors are not enough to take a treatment decision. Advances in our knowledge of bone matrix biochemistry, most notably of post-translational modifications in type I collagen, may allow identification of biochemical markers that reflect changes in the material property of bone, which is an important determinant of bone strength. Preliminary in vitro studies indicate that the extent of post-translational modifications of collagen--which can be reflected in vivo by the measurement of the urinary ratio between native and isomerised type I collagen--play a role in determining the mechanical competence of cortical bone, independently of BMD. Further studies in osteoporosis should explore the changes in these biochemical parameters of bone matrix as they may represent a key component of bone quality.

Biomarkers↗
PubMedDate not supplied

[Diagnosis of middle cerebral artery spasm by determination of flow velocity and the Lindegaard index with transcranial color Doppler sonography].

BACKGROUND AND PURPOSE: Recent reports have demonstrated the high utility of transcranial color-coded duplex sonography (TCCS) in the diagnosis of advanced spasm of the middle cerebral artery, whereas its accuracy in the diagnosis of mild vasospasm is reported to be lower. Relation of blood flow velocity in the middle cerebral artery (MCA) to that in the extracranial internal carotid artery (VMCA/VICA index) is recommended as being helpful in the diagnosis of vasospasm (the so called Lindegaard Index). Nevertheless, the exact diagnostic value of this index using the TCCS method remains to be established. The purpose of this study is to estimate the accuracy of TCCS in the diagnosis of MCA vasospasm, as based on the VMCA/VICA index. MATERIAL AND METHODS: The VMCA/VICA index was calculated in 195 patients (285 middle cerebral arteries) who were scheduled for cerebral arteriography. The TCCS study and color-coded duplex sonography of the internal carotid arteries were performed immediately before arteriography. RESULTS: A mild grade of MCA spasm was diagnosed angiographically in 21, and moderate-to-severe spasm in 29 MCAs out of 285 successfully insonated arteries. Peak-systolic, mean and end-diastolic blood velocities were measured using transcranial color sonography in the MCA and related to the respective velocities in the ipsilateral extracranial internal carotid artery. Receiver-operating characteristic curves (ROC) were calculated for particular velocities and for the related VMCA/VICA indices. By comparison of the areas under the ROC curves it was shown that the use of the VMCA/VICA index does not improve the accuracy of TCCS in the diagnosis of advanced MCA spasm, whereas it improves accuracy in the diagnosis of mild vasospasm. In particular, accuracy can be improved by the use of the VMCA/VICA index based on the end-diastolic velocity. The optimal diagnostic threshold of the VMCA/VICA index was determined at 3.9. CONCLUSIONS: The use of VMCA/VICA index improves the accuracy of TCCS in the diagnosis of mild vasospasm of the middle cerebral artery.

Adult↗
PubMed2005

[Osteoporosis].

The diagnosis of osteoporosis is based on the bone mineral density measurement using DXA technology, but the concept of bone quality mainly related to bone architecture is of great interest. The diagnostic threshold of osteoporosis is more and more modulated according to fracture risk factors such as age, prevalent fractures, bone turn-over leading to the definition of a therapeutic threshold. The use of bone resorption inhibitors before or with teriparatide decrease the response of the biochemical markers of bone formation and the gain in bone mineral density. The efficacy of strontium ranelate to decrease vertebral fracture incidence is confirmed and presented on non-vertebral fractures. Bisphosphonates are rapidly effective to prevent fractures and with a long term efficacy. Vitamin D treatment is effective to prevent falls.

Bone and Bones↗
PubMed2005

On caries prevalence and school-based fluoride programmes in Swedish adolescents.

Dental caries on approximal tooth surfaces in adolescents is still a problem in Sweden, as well as in many other industrialised countries. The aims of the present thesis were therefore: 1) to study whether caries prevalence is underestimated, 2) to evaluate the effect of cessation of fluoride mouth rinse (FMR) programme in schoolchildren with low caries prevalence, and 3) to reconsider the school as an arena for population-based fluoride (F) varnish and FMR programmes in order to minimise caries development in 13-16-year-olds. Paper I and II showed that the adolescents' own dentists consistently registered less caries compared to the recordings made by calibrated dentists from outside. Based on 420 16-year-olds at 12 different dental clinics in two neighbouring counties in Sweden, the differences were statistically significant at 10 of the 12 clinics when caries prevalence both was high in 1984 (Paper I) and somewhat lower in 1987 and 1990 (Paper II). About 80% of all approximal caries lesions were enamel lesions and are therefore not reported to the Swedish authorities, as only dentin lesions and fillings are currently included in the official caries data. The basis for Paper III was that many counties in Sweden abandoned school-based FMR programmes in the mid-late 1980s, as the official caries data revealed low caries prevalence among children. The cessation of FMR for 3 years for a group of 13-16-year-old adolescents (n=60) with low caries prevalence did not reveal any statistically significant differences in new caries lesions and fillings or in the progression of existing enamel lesions compared to a group of 13-16-year-olds (n=60) who continued to rinse for 3 years. Nor did a supplementary cross-sectional study reveal any differences in caries prevalence among adolescents who had or had not taken part in FMR programmes for the last 3 years. In Papers IV and V, new models for school-based F treatment were evaluated in two 3-year randomised controlled trials (RCT studies). Fluoride varnish (Duraphat) treatment, carried out at school by specially trained dental nurses using a simple mobile unit, among 854 13-16-year-olds from low, medium and high caries risk areas showed that the control groups developed more caries than the varnish groups (Paper IV). The largest difference was found in the high-risk area. The prevented fraction (PF) with approximal enamel lesions as the diagnostic threshold was 69% in high, 66% in medium and 20% in low risk areas for F varnish applied twice a year at six-month intervals. Supervised school-based FMR among 788 13-16-year-olds with low to moderate caries risk (Paper V) showed that FMR on the first three and the last three school days during the school term, i.e. 12 rinses/year, had a PF of 59%. Enamel lesions constituted > 90% of the new approximal caries lesions in both Papers IV and V. The main conclusions from this thesis are: (1) that enamel caries lesions on approximal surfaces should be included in epidemiological caries data in order to show true caries prevalence, (2) that the cessation of weekly FMR for 3 years did not lead to a statistically significant increase in caries incidence in a relatively small group of adolescents with low caries prevalence, and (3) that the school should be reconsidered as an arena for new models for population-based F school programmes. Fluoride varnish treatment twice a year at six-month intervals in medium and high caries risk areas, as well as supervised FMR on the first and the last three school days during the school term in low and medium caries risk areas, are excellent caries prevention strategies for approximal surfaces in adolescents at the caries risk ages of 13-16 years, as a supplement to dental home care and preventive efforts at dental clinics.

Adolescent↗
PubMed2005

Screening for blunt cerebrovascular injury: evaluating the accuracy of multidetector computed tomographic angiography.

BACKGROUND: Digital subtraction angiography (DSA) has been widely considered the "gold standard" for diagnosing blunt cerebral vascular injury, but recent advances in multidetector computed tomographic angiography (CTA) may have led to improved accuracy compared with DSA. METHODS: A retrospective study of patients who underwent CTA and DSA for suspected blunt cerebral vascular injury was performed. Two neuroradiologists and one radiology resident, blinded to the other test, independently reviewed CTA and DSA studies. Sensitivity and specificity of CTA were determined using DSA as the gold standard. RESULTS: Thirty-two patients met the study criteria. Consensus DSA interpretation identified 17 vascular injuries in 15 patients. Among the three radiologists, the sensitivity of CTA for carotid injury ranged from 83% (10 of 12) to 92% (11 of 12), and the specificity ranged from 88% (36 of 41) to 98% (40 of 41). The sensitivity for vertebral artery injury ranged from 40% (2 of 5) to 60% (3 of 5), and the specificity ranged from 90% (35 of 39) to 97% (38 of 39). CTA sensitivity rose to 80% (four of five) in the vertebral artery when a lower diagnostic threshold was used. CONCLUSION: The sensitivity of CTA for carotid artery injury at our center was higher than in previous studies. This might be attributable to the superior quality of multidetector CTA images. Our conclusions on CTA accuracy in the vertebral artery are limited by the small number of injuries.

Adolescent↗
PubMed2006

Insulin resistance & secretion in subjects with normal fasting plasma glucose.

BACKGROUND & OBJECTIVES: Lowering the diagnostic threshold of normal fasting plasma glucose (FPG) from 6.1 to 5.56 mmol/l has been proposed by American Diabetes Association. As early detection of subjects with risk of diabetes is important, it is crucial to understand the relationship between different levels of FPG, insulin resistance (IR) and insulin secretion. We measured IR and insulin secretion in subjects with different levels of FPG to investigate the relationship between FPG and IR and insulin secretion, and to see whether lowering the cut-off improves the sensitivity of impaired fasting glucose to predict diabetes. METHODS: Apparently healthy subjects (294) were enrolled and divided into 3 groups according to their FPG levels; group 1 (<5.56 mmol/l, n=129), group 2 (5.56-6.09 mmol/l, n=99) and group 3 (6.1-6.9 mmol/l, n=66). Steady state plasma glucose (SSPG) derived from insulin suppression test was used to quantify insulin resistance and 30 min insulinogenic index during an oral glucose tolerance test to measure acute insulin secretion. RESULTS: The blood pressure was higher in group 3 than group 2 and fasting plasma insulin and triglyceride were higher in group 3 than that in group 1. The insulinogenic index was higher in the group 1 than group 3 and the SSPG was significantly higher in group 3 compared to group 1. INTERPRETATION & CONCLUSION: Our study showed a trend of progressive deterioration of both insulin action and secretion with increasing FPG level. These are supportive evidences for lowering the FPG to 5.56 mmol/l as suggested by the ADA.

Adult↗
PubMed1991

Use of augmented decision tables to convert probabilistic data into clinical algorithms for the diagnosis of appendicitis.

Decision table techniques have been shown to be useful for ensuring logical completeness, eliminating ambiguity, and optimizing the translation of logic into flowcharts or computer programs. Nevertheless, they have not been widely applied in medicine. We have used decision table techniques to demonstrate the derivation of two sets of rules for determining whether to operate on patients with suspected appendicitis based on patterns of observed signs and symptoms. One rule set is based on a diagnostic threshold whereby morbidity is minimized; the other rule set minimizes mortality. For this purpose, we have developed an augmented decision table format that allows the incorporation of probability and utility data.

Algorithms↗
PubMed1994

Problems in diagnosing nosocomial pneumonia in mechanically ventilated patients: a review.

OBJECTIVE: To review the available information about diagnosing nosocomial pneumonia in mechanically ventilated intensive care unit patients. DATA SOURCES: The National Library of Medicine MEDLINE database and bibliographies of the reviewed articles. STUDY SELECTION AND DATA EXTRACTION: Studies were selected from the English literature, with an emphasis on recent studies. Studies were selected on the basis of their historical value and originality. A total of 82 articles was considered relevant. DATA SYNTHESIS: Many efforts have been made to diagnose nosocomial pneumonia with more accuracy. Two promising new diagnostic modalities are the protected specimen brush and bronchoalveolar lavage, both performed via flexible bronchoscopy. Depending on the study design and patient selection, sensitivity and specificity of bronchoalveolar lavage and protected specimen brush differ widely, ranging from 50% to 100%. However, interpretation of study results is seriously hampered because a standardized diagnostic threshold is not available, the influence of previous antimicrobial therapy on culture results remains unclear, and, most importantly, a well-defined gold standard to study the value of both methods is lacking. CONCLUSIONS: Because studies of bronchoalveolar lavage and protected specimen brush are seriously hampered by several pitfalls, these studies as well as clinical results must be interpreted with caution.

Bronchoalveolar Lavage Fluid↗
PubMed1994

[The use of myoglobin for the early diagnosis of reperfusion in acute myocardial infarct].

Myoglobin (Mb) is an intracardiac protein known to be rapidly released after onset of reperfusion. The new automated latex-enhanced nephelometric immunoassay is an interesting fast alternative to the other methods. Therefore we sought to assess the accuracy of rapid rise of Mb concentration to predict the success of reperfusion. Thirty consecutive patients, admitted to the coronary care unit for first acute Q-wave myocardial infarction, were prospectively studied. They underwent thrombolysis 126 +/- 45 min (range 30-180) after onset of symptoms. Mb determinations were obtained before starting therapy and, thereafter, hourly for 5 hours, and after 8 and 12 hours. The usual non invasive markers of reperfusion were considered, and coronary arteriography was performed in all patients at 5.5 +/- 0.9 days (range 1-6). In 23 patients (77%) patent infarct-related artery (IRA) (TIMI II-III) and in 7 (23%) occluded IRA (TIMI 0-I) was observed at coronary arteriography. We modified the judgment in 3 cases (10%) with occluded IRA. All reperfused patients (Group A) showed a rapid rise of Mb concentration with a mean time to peak occurring 133 +/- 80 min (range 60-180) after therapy. Mean value of unreperfused patients (Group B) was 330 +/- 173 min (range 180-420). We retrospectively stated a time to peak of Mb concentration within 180 min as diagnostic threshold for reperfusion. Sensitivity was 100%, specificity 50%, positive predictive value 93%, negative predictive value 100%. The results of this study confirm that Mb release kinetics may play an interesting role in the early prediction of patency of IRA in acute myocardial infarction after thrombolysis.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1997

An overview and ranking of biochemical markers of cardiac disease. Strengths and limitations.

Comparing the performance of one biochemical marker of myocardial damage with another is bedeviled with many problems, including a lack of uniform calibration between instruments providing measurements of the same analyte and differing reference ranges and decision thresholds. Diagnostic groupings and various gold standards create difficulties in comparing analytic results. There is a lack of uniformity in test assessments and their use for diagnostic and prognostic purposes. Finally, there is an almost complete absence of data from economic and decision analyses of test usage. It is argued that a much more stringent and rigorous approach to these aspects is essential. The use of biochemical markers of cardiac damage is in a dynamic state, with new applications continually appearing and new markers being developed. It is therefore essential that a uniform and rigorous outlook be maintained to ensure both optimal and economic test utilization based on the previously outlined principles.

Biomarkers↗
PubMed1998

Transferrin receptor assay and zinc protoporphyrin as markers of iron-deficient erythropoiesis in end-stage renal disease patients.

To evaluate storage iron deficiency and iron-deficient erythropoiesis we determined, in a cross-sectional study of 95 patients mainly including end-stage renal disease patients (ESRD) with (32) and without rh-EPO therapy (55), the following parameters: hemoglobin, mean corpuscular red cell volume, ferritin, transferrin saturation (TS), zinc protoporphyrin (ZPP) and soluble transferrin receptor (TfR). In the dialysis group the percentage of positive samples with each marker of tissue iron supply defined as TS < 20%, ZPP > 40 mumol/mol Heme and TfR > 3.05 microgram/ml was as follows: TS 43.7% and 32.2% at a diagnostic threshold level of < 16%, ZPP 33.3% and TfR 17.2%. Manifest storage iron deficiency defined as ferritin < 30 ng/ml was observed in 5.7% of the samples while the mean ferritin concentration of the rh-Epo treated dialysis patients was 509.3 ng/ml compared to 262.5 ng/ml in the group without rh-EPO therapy. These data reflect a generous iron substitution in our series taking a TS < 20% as an intervention criterion. Looking at the different results of the three markers the best correspondence was found between ZPP and TfR resulting in a weak positive correlation (+0.64). In conclusion, we found quite different results with different assays when evaluating endogenous iron availability in our series of mainly ESRD patients in a cross-sectional study. Because a gold-standard is not defined further firm conclusions cannot be drawn from this type of study. The adequacy of the different parameters of iron metabolism including threshold levels and, consequently, the decision and route of iron substitution deserve an evaluation in a longitudinal study to characterize the best marker or marker combination in this setting.

Adult↗
PubMed1998

Prevalence of autoantibodies to thyroglobulin in dogs with nonthyroidal illness.

OBJECTIVE: To evaluate a thyroglobulin autoantibody (TgAA) assay and determine a diagnostic threshold. SAMPLE POPULATION: Serum samples from dogs with various endocrine abnormalities and from 30 obese adult female Beagles. PROCEDURE: TgAA were determined by use of the ELISA. Six experiments were done: 1, definition of positive results for TgAA using samples from normal and T3 autoantibody (T3AA) positive dogs; 2, establishment of prevalence of positive results in 91 clinically normal dogs; 3, evaluation of positive results for sera from dogs with nonthyroidal illnesses; 4, testing of samples from dogs with primary hypothyroidism but absence of T4AA or T3AA, or both; 5, determination of prevalence of false-negative results in dogs that are T4AA and/or T3AA positive, which were (18 dogs) or were not (22 dogs) receiving L-thyroxine replacement therapy; and 6, examination of thyroid biopsy specimens from 18 dogs (8 TgAA positive and 10 TgAA negative). RESULTS: Positive results were defined as at least twice (200%) the optical density of the negative-control sample. False-positive results were obtained for only 3.4% of 146 dogs with nonthyroidal illness. Thirty-seven percent of dogs with primary hypothyroidism, but no evidence of T4AA or T3AA, or both, were TgAA positive. False-negative results were found in 1 of 22 and 2 of 18 T3AA-positive dogs with and without thyroid replacement therapy, respectively. Thyroid biopsy specimens from 8 TgAA-positive dogs had evidence of lymphocytic thyroiditis, whereas those from 10 TgAA-negative dogs did not. CONCLUSION AND CLINICAL RELEVANCE: The assay is sensitive and specific for identification of lymphocytic autoimmune thyroiditis in dogs, and has potential for aiding early diagnosis of thyroiditis in dogs and identifying dogs likely to perpetuate hypothyroidism in breeding programs.

Animals↗
PubMed1997

[Evaluation of electrophysiological diagnosis of concealed accessory pathway (CAP) during transesophageal atrial pacing (TEAP)].

CAP were diagnosed by TEAP using RPE intervals in 55 cases of paroxysmal supraventricular tachycardia (PSVT). There were no obvious CAP manifestations of ECG in those cases, the mean RPE interval during tachycardia was 140 +/- 29 ms, and the lower limit of x +/- 2 s was 82 ms. Among them, the VA intervals were measured by electrocardiophysiological examination in 21 cases. The mean value was 130 +/- 25 ms, and the lower limit of x +/- 2 s was 80 ms. The results suggest that RPE over 82 ms may be an important diagnostic threshold cut off point of atrioventricular reciprocation.

Adolescent↗
PubMedDate not supplied

Threshold analysis using diagnostic tests with multiple results.

Clinical problems represented by decision trees can be analyzed in terms of the probability threshold model, which provides management recommendations based on the prior probability of disease, the test threshold, and the test-treatment threshold. As originally proposed, the threshold model assumes that diagnostic tests provide information about a single event that is relevant to the decision. For some problems, however, a diagnostic test may provide information about more than one such event (e.g., a computed tomography [CT] scan gives information about both mediastinal and hilar metastases in lung cancer). The authors extend the probability threshold model to cases in which a single test provides information about two events that are relevant to the decision. They derive four thresholds that determine the best strategy for any combination of test results. The approach is illustrated for the decision to use a CT scan to stage lung cancer. The analysis reveals that: 1) the range of prior probabilities for which testing is optimal increases; 2) for some prior probabilities only test results about one event are important; 3) for some prior probabilities test results about both events are important; and 4) failure to account fully for information provided by a test can lead to erroneous test and treatment recommendations.

Bayes Theorem↗
PubMed1983

Derived thresholds. Determining the diagnostic probabilities at which clinicians initiate testing and treatment.

The medical decision making literature has previously considered the test and test-treatment thresholds in a normative fashion. In the normative approach, the analyst calculates the optimal threshold--the likelihood of disease at which testing or treatment should be undertaken. In contrast, we describe a method of deriving the threshold in a descriptive fashion, by determining the probabilities of disease at which clinicians actually make the decision to test or to initiate specific treatment without further testing. In applying this method, the analyst first asks clinicians to provide an estimate of the prior probability of disease, and to select one of three options: test, treat, or do neither. After receiving new information about the patient, the clinicians are asked to revise the probability estimate and to select a new option. Correlation of changes in the probability of a disease with changes in the clinicians' selections of options to test or treat enables the analyst to estimate the test and test-treatment thresholds used by the clinicians in medical decision making. Knowledge of these thresholds also enables the analyst to calculate the clinicians' ratio of the benefits to the costs of the therapy being considered, considering the risks of the test itself.

Cost-Benefit Analysis↗