Identifying and managing hereditary risk of breast and ovarian cancer.

In the past, all women with a family history of breast or ovarian cancer were considered to be at increased risk of cancer themselves. The discovery of BRCA1 and BRCA2 demonstrated that susceptibility to breast and ovarian cancer can be inherited by women as a single-gene autosomal dominant disorder. For such women, evaluation of family history is an important screening tool to identify the possibility of hereditary cancer risk but only genetic testing can provide definitive, individualized risk assessment. Women who have inherited mutations in BRCA1 or BRCA2 now have several medical management options to address their increased risk of cancer. A well-educated community of health care providers and patients can use hereditary risk assessment, including genetic testing, to improve health care.

The predictive value of BRCA1 and BRCA2 mutation testing.

Genetic testing for mutations in BRCA1 and BRCA2, two genes predisposing to breast and ovarian cancers, is available to women with a relevant family history. The aim of this study was to estimate the positive and negative predictive value of clinical sequence analysis of these genes. A reference graph showing positive and negative predictive values over a range of pre-test risk was derived, taking into account the sensitivity and specificity of a full-sequence analysis test. High positive and negative predictive values were found for women with pre-test risk between 4% and 40%, a range of risk commonly seen in clinical testing. The predictive value of full sequence and single-site analysis of BRCA1 and BRCA2, therefore, compares favorably with other diagnostic medical tests. Our results provide a numerical estimate of the predictive value of BRCA testing, and as such, provide a valuable tool to healthcare providers and families as they interpret BRCA1 and BRCA2 test results.

Calibration verification of the International Normalized Ratio.

The International Normalized Ratio (INR) system for reporting the prothrombin time (PT) is essentially a calibration activity intended to standardize PT reporting across various reagent/instrument systems. However, complete standardization of PT reporting through the INR has been difficult to achieve for a variety of reasons, including inaccurate assignment of thromboplastin International Sensitivity Indexes (ISIs) and specific (local) reagent/instrument effects. Until now, the individual laboratory has not been able to easily verify the accuracy of its INR. Using standard lyophilized plasmas with INR values assigned against IRP RBT/90 rabbit thromboplastin, the authors present a method that allows a laboratory to locally verify its range of accuracy for the INR. The method is illustrated on a single coagulometer with two thromboplastin lots of differing sensitivity (Pacific Hemostasis Thromboplastin-DS and Thromboplastin-D from rabbit sources, with respective International Sensitivity Indexes of 1.20 and 1.97). In this illustration of the method, the accuracy of Thromboplastin-DS was superior to that of Thromboplastin-D. Interpretation of the data and cautions regarding the use of standard plasmas for calibration verification are discussed. Using this method, a reportable range of accuracy at a given error tolerance can be established locally for INR measurements within a laboratory. Laboratories of any size can apply this method to study the accuracy of their INR reagent/instrument systems, thus performing calibration verification. When used in conjunction with assessments of assay precision, this method can help laboratories to select better reagent/instrument systems and thereby produce more accurate and more clinically meaningful INR results.

The influence of the reference mean prothrombin time on the international normalized ratio.

The International Normalized Ratio (INR) is a mathematical transformation of the prothrombin time (PT). The transformation requires a laboratory to compute the geometric mean of its own reference population. In this paper, the authors examine how the reference mean PT influences the INR accuracy and precision using a validated probabilistic model. The variance of the geometric mean of reference populations in three laboratory settings was determined. Because the variance of an individual laboratory geometric mean is not directly determinable by simple parametric equations, its variance is estimated using bootstrap analysis. The geometric mean is compared to the computationally simpler arithmetic mean for effects on accuracy and precision of the resulting INR. The study shows mathematically and empirically that using the arithmetic mean biases INR determinations so that patients tend to be over-anticoagulated. However, in the laboratory settings examined, the amount of bias was both statistically and clinically insignificant. An analysis of the effect on the INR of errors in estimating the geometric mean reference PT also is performed. For large biases in estimating the reference mean, the INR can be significantly affected and can trigger inappropriate clinical actions in patients. The authors demonstrate empirically and mathematically that biases in the geometric mean reference PT do not affect the INR coefficient of variation. However, they produce significant differences in confidence intervals for INR determinations. Laboratories must exercise care in determining specific reference means to ensure that biases do not occur in geometric mean reference PT determinations. This can be achieved by circumspection in the selection of normal subjects for the reference population, carefully reviewing the data, and performing the proper calculations on the data.

The international normalized ratio and uncertainty. Validation of a probabilistic model.

The motivation behind the creation of the International Normalized Ratio (INR) was to improve interlaboratory comparison for patients on anticoagulation therapy. In principle, a laboratory that reports the prothrombin time (PT) as an INR can standardize its PT measurements to an international reference thromboplastin. Using probability theory, the authors derived the equation for the probability distribution of the INR based on the PT, the International Sensitivity Index (ISI), and the geometric mean PT of the reference population. With Monte Carlo and numeric integration techniques, the model is validated on data from three different laboratories. The model allows computation of confidence intervals for the INR as a function of PT, ISI, and reference mean. The probabilistic model illustrates that confidence in INR measurements degrades for higher INR values. This occurs primarily as a result of amplification of between-run measurement errors in the PT, which is inherent in the mathematical transformation from the PT to the INR. The probabilistic model can be used by any laboratory to study the reliability of its own INR for any measured PT. This framework provides better insight into the problems of monitoring oral anticoagulation.

Examination of International Normalised Ratio (INR) imprecision by comparison of exact and approximate formulas. Intermountain Laboratory Data Project.

AIM: To evaluate the accuracy of the approximate linear formula of International Normalised Ratio (INR) imprecision by formal mathematical analysis. METHODS: Using probability theory, an exact formula for the coefficient of variation (CV) of the INR was derived. The CV from the approximate formula was compared with the CV from the exact formula for INR determinations between 1.0 and 10.0 with International Sensitivity Indices (ISIs) between 1.0 and 3.0 and prothrombin time ratio CVs between 1.0 and 10.0%. RESULTS: When the ISI equals 1.0, the approximate formula and the exact formula are equal. When the ISI is more than 1.0, the approximate formula overestimates the exact CV, but by less than one hundredth of the exact CV in the parameter ranges studied. The approximate formula is most accurate when laboratories achieve excellent prothrombin time measurement precision and use sensitive thromboplastins. CONCLUSIONS: The approximate formula provides a simple means for estimating the imprecision of the INR and is sufficiently accurate to warrant its use in clinical laboratories.

Antibody-mediated interference of a homogeneous immunoassay.

We hypothesized that an antibody-mediated interference could arise in a homogeneous immunoassay used to determine the presence of cocaine metabolites in urine. Urine specimens containing benzoylecgonine (BE) at concentrations near the National Institute on Drug Abuse (NIDA) threshold were assayed in replicate determinations by EMIT. Excess reagent protein (containing antibody specific for cocaine metabolites) was added to specimens to test for an antibody-mediated interference. Replicates of the BE-fortified specimens tested by EMIT that did not contain excess reagent antibody were all positive by the assay, while those that contained the excess reagent antibody were all negative. Because it may be difficult to detect excess interfering antibody by using some traditional tests for urine adulteration, we present these findings to illustrate a potential problem for some homogeneous immunoassays in forensic urine drug testing programs.

Nonparametric assessment of toxicologic assay linearity by bootstrap analysis.

An important aspect in the evaluation of toxicologic assay methodology is the assessment of calibration. This paper presents a new approach for validating calibration using bootstrap analysis. The technique is illustrated with a quantitative assay for benzoylecgonine in urine by gas chromatography/mass spectrometry (GC/MS). Because the bootstrap analysis does not rely on parametric assumptions regarding the distribution of errors about the regression line, it can be applied to situations where the parametric distribution of response errors about the regression line is unknown. This application of bootstrap methodology yields a probabilistic measure of confidence on the linearity of the calibration curve.

Parental perceptions of genetic risk: correlation with choice of prenatal diagnostic procedures.

Determinants of patients' decisions to have prenatal diagnosis, and by which test (amniocentesis or chorionic villus sampling), is not well understood. We have administered questionnaires to couples referred for genetic counseling and procedures for advanced maternal age and determined patients' responses, decisions, and attributable determinants of their choice. Differences in perception of risks between partners, and choice of procedure, correlated with socioeconomic, educational status and age.

A confidence profile analysis of the results of femoropopliteal percutaneous transluminal angioplasty in the treatment of lower-extremity ischemia.

The many reports on the use of percutaneous transluminal angioplasty (PTA) for treating lower-extremity ischemia were analyzed to determine the outcomes of femoropopliteal PTA used in patients with intermittent claudication (IC) and with more severe limb-threatening ischemia ("salvage"). The Confidence Profile Method was used to interpret and adjust the evidence from 12 selected clinical reports, and to combine the data into single best estimates of the outcomes that were considered important for decision making. The following combined estimates were obtained: The early success rate was 89% +/- 2.5% after PTA for IC, compared with 77% +/- 4% after PTA for salvage. The largest decline in patency occurred during the first 6 to 12 months. After 2 years there was little further attrition. Three-year patency was 62% +/- 9% for IC and 43% +/- 7% for salvage. The risk of procedure-related mortality and limb loss was low, estimated at 0.5% and 1%, respectively. The risk of serious complications requiring surgical repair was about 2.5%. The Confidence Profile Method is a novel way of obtaining this information from the imperfect clinical literature. The method incorporates subjective judgments and assumptions and gives quantitative and visual representations of both the magnitude of the estimated outcome and the uncertainty associated with it. The results are as strong as the assumptions incorporated in them are valid. The estimates derived in this study can be used for any decision on the use of PTA in the femoropopliteal segment.

Probabilistic sensitivity analysis methods for general decision models.

Probabilistic sensitivity analysis has previously been described for the special case of dichotomous decision trees. We now generalize these techniques for a wider range of decision problems. These methods of sensitivity analysis allow the analyst to evaluate the impact of the multivariate uncertainty in the data used in the decision model and to gain insight into the probabilistic contribution of each of the variables to the decision outcome. The techniques are illustrated using Monte Carlo simulation on a trichotomous decision tree. Application of these powerful tools permits the decision analyst to investigate the structure and limitations of more complex decision problems with inherent uncertainties in the data upon which the decisions are based. Probabilistic sensitivity measures can provide guidance into the allocation of resources to resolve uncertainty about critical components of medical decisions.

Probabilistic analysis of decision trees using Monte Carlo simulation.

The authors describe methods for modeling uncertainty in the specification of decision tree probabilities and utilities using Monte Carlo simulation techniques. Exact confidence levels based upon the underlying probabilistic structure are provided. Probabilistic measures of sensitivity are derived in terms of classical information theory. These measures identify which variables are probabilistically important components of the decision. These techniques are illustrated in terms of the clinical problem of anticoagulation versus observation in the setting of deep vein thrombosis during the first trimester of pregnancy. These methods provide the decision analyst with powerful yet simple tools which give quantitative insight into the structure and inherent limitations of decision models arising from specification uncertainty. The techniques may be applied to complex decision models.

Probabilistic analysis of decision trees using symbolic algebra.

Uncertainty in medical decision making techniques occurs in the specification of both decision tree probabilities and utilities. Using a computer-based algebraic approach, methods for modeling this uncertainty have been formulated. This analytic procedure allows an exact calculation of the statistical variance at the final decision node using automated symbolic manipulation. Confidence and conditional confidence levels for the preferred decision are derived from gaussian theory, and the mutual information index that identifies probabilistically important tree variables is provided. The computer-based algebraic method is illustrated for a problem previously analyzed by Monte Carlo simulation. This methodology provides the decision analyst with a procedure to evaluate the outcome of specification uncertainty, in many decision problems, without resorting to Monte Carlo analysis.

Improved reference-interval estimation.

We used two standard methods (the common percentile and the log-power gaussian transformation) and two novel methods (weighted percentile and smoothed spline interpolation) to estimate the 2.5th and 97.5th percentiles of 1000 sets of data from eight diverse distributional forms, generated by Monte Carlo simulation. For each distributional form we derived an estimate of optimal performance. Although none of the four proposed methods closely approximated the optimal performance bound, the spline interpolation method and the weighted percentile method were superior to the two standard methods in accurately estimating percentiles.

Automatic prediction of platelet utilization by time series analysis in a large tertiary care hospital.

The authors investigated the potential of various time series models to aid in the prediction of platelet utilization patterns at the authors' institution. During a seven-week feasibility study, prediction accuracy of time series models was slightly inferior to that of an experienced platelet transfusion coordinator. The models that adjusted for day-of-the-week variability were superior to simpler models in a prospective evaluation over a three-month period. Conversion to a time series model for platelet utilization prediction has resulted in significant savings in personnel costs for management of platelet concentrate inventory in the blood bank. Outdate rates have been lower than those achieved by the platelet transfusion coordinator, with no sacrifice in availability occurring during the evaluation period.