Search PubMed⌕ Search

Biomedical subjects

D B Matchar

Publications and source records attributed to D B Matchar.

98 records · Page 6Linked to original sources
PubMedDate not supplied

Predicting the outcomes of electrophysiologic studies of patients with unexplained syncope: preliminary validation of a derived model.

PURPOSE: To develop and validate a predictive model that would allow clinicians to determine whether an electrophysiologic (EP) study is likely to result in useful diagnostic information for a patient who has unexplained syncope. PATIENTS: One hundred seventy-nine consecutive patients with unexplained syncope who underwent EP studies at two university medical centers comprised the training sample. A test sample to validate the model was made up of 138 patients from the clinical literature who had undergone EP studies for syncope. DESIGN: Retrospective analysis of patients undergoing EP studies for syncope. The data collector was blinded to the study hypothesis; the electrophysiologist assessing outcomes was blinded to clinical and historical data. Clinical predictor variables available from the history, the physical examination, electrocardiography (ECG), and Holter monitoring were analyzed via two multivariable predictive modeling strategies (ordinal logistic regression and recursive partitioning) for their abilities to predict the results of EP studies, namely tachyarrhythmic and bradyarrhythmic outcomes. These categories were further divided into full arrhythmia and borderline arrhythmia groups. RESULTS: Important outcomes were 1) sustained monomorphic ventricular tachycardia (VT) and 2) bradyarrhythmias, including sinus node and atrioventricular (AV) conducting disease. The results of the logistic regression (in this study, the superior strategy) showed that the presence of organic heart disease [odds ratio (OR) = 3.0, p less than 0.001] and frequent premature ventricular contractions on ECG (OR = 6.7, p less than 0.004) were associated with VT, while the following abnormal ECG findings were associated with bradyarrhythmias: first-degree heart block (OR = 7.9, p less than 0.001), bundle-branch block (OR = 3.0, p less than 0.02), and sinus bradycardia (OR = 3.5, p less than 0.03). Eighty-seven percent of the 31 patients with important outcomes at EP study had at least one of these clinical risk factors, while 95% of the patients with none of these risk factors had normal or nondiagnostic EP studies. In the validation sample, the presence of one or more risk factors would have correctly identified 88% of the test VT patients and 65% of the test bradyarrhythmia patients as needing EP study. CONCLUSION: These five identified predictive factors, available from the history, the physical examination, and the initial ECG, could be useful to clinicians in selecting those patients with unexplained syncope who will have a serious arrhythmia identified by EP studies.

Arrhythmias, Cardiac↗
PubMedDate not supplied

Intermediate, indeterminate, and uninterpretable diagnostic test results.

Diagnostic tests do not always yield positive or negative results; sometimes the results are intermediate, indeterminate, or uninterpretable. No consensus exists for the incorporation of such results into data assessment. Conventional Bayesian analysis leads investigators to either exclude patients with non-positive, non-negative results from their studies or categorize such results into inappropriate cells of the standard four-cell decision matrix. The authors propose a standardized method for reporting results in studies dealing with diagnostic test use and discuss how researchers should expand the four-cell matrix to six cells when non-positive, non-negative results occur. They suggest that the six-cell matrix with new operational definitions of sensitivity, specificity, likelihood ratios, and test yield should be adopted routinely. In addition, they define the different types of non-positive, non-negative results and demonstrate how clinicians can use tree-structured decision analysis from the six-cell matrix. While their method does not solve all problems posed by non-positive, non-negative results, it does suggest a standard method for reporting these results and utilizing all the data in decision making.

Bayes Theorem↗
PubMedDate not supplied

A Bayesian method for evaluating medical test operating characteristics when some patients' conditions fail to be diagnosed by the reference standard.

The evaluation of a diagnostic test when the reference standard fails to establish a diagnosis in some patients is a common and difficult analytical problem. Conventional operating characteristics, derived from a 2 x 2 matrix, require that tests have only positive or negative results, and that disease status be designated definitively as present or absent. Results can be displayed in a 2 x 3 matrix, with an additional column for undiagnosed patients, when it is not possible always to ascertain the disease status definitively. The authors approach this problem using a Bayesian method for evaluating the 2 x 3 matrix in which test operating characteristics are described by a joint probability density function. They show that one can derive this joint probability density function of sensitivity and specificity empirically by applying a sampling algorithm. The three-dimensional histogram resulting from this sampling procedure approximates the true joint probability density function for sensitivity and specificity. Using a clinical example, the authors illustrate the method and demonstrate that the joint probability density function for sensitivity and specificity can be influenced by assumptions used to interpret test results in undiagnosed patients. This Bayesian method represents a flexible and practical solution to the problem of evaluating test sensitivity and specificity when the study group includes patients whose disease could not be diagnosed by the reference standard.

Algorithms↗
PubMedDate not supplied

Global judgments versus decision-model-facilitated judgments: are experts internally consistent?

A widely used method for evaluating the appropriateness of medical procedures and practices is the "modified Delphi" approach using expert panelists' global ratings. However, several difficulties in the assignment of global ratings have led to a search for alternative methods, including the use of decision models. To examine the potential impact of using decision models with an expert panel, the authors compared a panel's global ratings for the appropriateness of carotid endarterectomy with the results of a decision-analytic model in which expert panelists estimated probabilities and utilities that were used as inputs for the model. For 17 different patient scenarios, the nine expert panelists showed variability in "calibration" between the two methods, with their expected utilities calculated from the model generally being higher than their global ratings. However, the correlation between the two methods was excellent. When the panel's median global utility was compared with the panel's median expected utility calculated from the model, the Spearman correlation coefficient was 0.88. This study demonstrated that an expert panel's appropriateness ratings and their expected utilities were highly correlated. In addition, the panelists appeared to be internally consistent in that their judgments about individual probabilities and utilities were correlated with their global judgments. These results should encourage additional efforts to incorporate decision models into the process of clinical guideline development. The authors believe that decision models can help improve a panel's capacity to understand and reconcile discordance, and increase their satisfaction that the process reflects the best possible judgments.

Carotid Stenosis↗
PubMedDate not supplied

Assessing uncertainty in cost-effectiveness analyses: application to a complex decision model.

A framework for quantifying uncertainty about costs, effectiveness measures, and marginal cost-effectiveness ratios in complex decision models is presented. This type of application requires special techniques because of the multiple sources of information and the model-based combination of data. The authors discuss two alternative approaches, one based on Bayesian inference and the other on resampling. While computationally intensive, these are flexible in handling complex distributional assumptions and a variety of outcome measures of interest. These concepts are illustrated using a simplified model. Then the extension to a complex decision model using the stroke-prevention policy model is described.

Bayes Theorem↗
PubMedDate not supplied

Predicting the cost of illness: a comparison of alternative models applied to stroke.

Predictions of cost over well-defined time horizons are frequently required in the analysis of clinical trials and social experiments, for decision models investigating the cost-effectiveness of interventions, and for macro-level estimates of the resource impact of disease. With rare exceptions, cost predictions used in such applications continue to take the form of deterministic point estimates. However, the growing availability of large administrative and clinical data sets offers new opportunities for a more general approach to disease cost forecasting: the estimation of multivariable cost functions that yield predictions at the individual level, conditional on intervention(s), patient characteristics, and other factors. This raises the fundamental question of how to choose the "best" cost model for a given application. The central purpose of this paper is to demonstrate how to evaluate competing models on the basis of predictive validity. This concept is operationalized according to three alternative criteria: 1) root mean square error (RMSE), for evaluating predicted mean cost; 2) mean absolute error (MAE), for evaluating predicted median cost; and 3) a logarithmic scoring rule (log score), an information-theoretic index for evaluating the entire predictive distribution of cost. To illustrate these concepts, the authors conducted a split-sample analysis of data from a national sample of Medicare-covered patients hospitalized for ischemic stroke in 1991 and followed to the end of 1993. Using test and training samples of about 500,000 observations each, they investigated five models: single-equation linear models, with and without log transform of cost; two-part (mixture) models, with and without log transform, to directly address the problem of zero-cost observations; and a Cox proportional-hazards model stratified by time interval. For deriving the predictive distribution of cost, the log transformed two-part and proportional-hazards models are superior. For deriving the predicted mean or median cost, these two models and the commonly used log-transformed linear model all perform about the same. The untransformed models are dominated in every instance. The approaches to model selection illustrated here can be applied across a wide range of settings.

Cerebrovascular Disorders↗
PubMedDate not supplied

Life-sustaining therapy. A model for appropriate use.

New strategies are needed to curb the proliferation of life-sustaining therapies that rarely benefit patients. We propose a model for appropriate use of such therapies that incorporates effectiveness, utility, and marginal costs. If a therapy is rarely effective and rarely desirable, it is considered medically inappropriate. If the marginal cost-effectiveness ratio is inordinately high, it is considered economically inappropriate. If a therapy is either medically or economically inappropriate, it should not be automatically offered. The model provides an operational definition of futility and is illustrated with an analysis of out-of-hospital cardiopulmonary resuscitation for chronically ill older people. Advance directives, explicit health care rationing, and defining futile therapy based on survival predictions are alternatives to the appropriate care model, but are insufficient strategies to solve the problem of inappropriate life-sustaining care.

Aged↗