Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Bayesian modelling”

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 1,657 records · Page 92Linked to original sources
PubMed2006

Major advances in genetic evaluation techniques.

The past quarter-century in genetic evaluation of dairy cattle has been marked by evolution in methodology and computer capacity, expansion in the array of evaluated traits, and globalization. Animal models replaced sire and sire-maternal grandsire models and, more recently, application of Bayesian theory has become standard. Individual test-day observations have been used more effectively in estimation of lactation yield or directly as input to evaluation models. Computer speed and storage are less limiting in choosing procedures. The increased capabilities have supported evaluation of additional traits that affect the net profitability of dairy cows. The importance of traits other than yield has increased, in a few cases due to an antagonistic relationship with yield. National evaluations combined internationally provide evaluations for bulls from all participating countries on each of the national scales, facilitating choices from among many more bulls. Selection within countries has increased inbreeding and the use of similar genetics across countries reduces the previously available outcross population. Concern about inbreeding has prompted changes in evaluation methodology and mating practices, and has promoted interest in crossbreeding. In just the past decade, distribution of genetic evaluations has gone from mailed paper or computer tapes for a limited audience to publicly accessible, request-driven distribution via the Internet. Among the distributed information is a choice of economic indices that combine an increasing array of traits into numbers reflecting breeding goals under different milk-pricing conditions. Considerable progress in genomics and the mapping of the bovine genome have identified markers for some deleterious recessive genes, but broader benefits of marker-assisted selection are still in the future. A possible exception is the proprietary use of DNA testing by semen producers to select among potential progeny test bulls. The collection and analysis of industry-wide data to evaluate genetic merit will continue to be the most important tool for genetic progress into the foreseeable future.

Animals↗
PubMed1997

Comparative Analysis of Coronary Surgery Risk Stratification Models.

BACKGROUND: Preoperative risk assessment models for coronary bypass surgery (CABG) have been proposed, but comparison of them using independent databases needs to be done. METHODS: Models of CABG hospital mortality were tested on a set of 3,443 patients who underwent CABG including a subset of 3,237 patients who had isolated CABG (no valve procedures), in our database since 1991. Four models previously described were designated as Parsonnet (PS), Cleveland (CL), and Society of Thoracic Surgeons version 1 (ST1) and version 2 (ST2). We developed our own Bayesian (BA) and logistic regression (LR) models and calibrated the PS and CL models on 2,842 patients operated on prior to 1991. Models were compared with respect to 1) mean predicted mortality, 2) correlation of predicted to observed mortality, 3) Brier mean probability score, 4) descriptive statistics, 4) the C-Index (area beneath the receiver operating characteristic curve), and 5) predictive efficiency. Since the ST1 and ST2 models were developed for use only with isolated CABG patients, these models were compared with the others using an isolated CABG subset. RESULTS: Observed mortality for all 3,443 CABG patients was 4.0%. For this group, the mean mortality predicted by PS, CL, BA, LR, was 9.0 +/- 8.0, 6.0 +/- 6.0, 7.6 +/- 15.6, and 5.1 +/- 7.7 (mean +/- standard deviation) respectively. C-Indexes were.80 +/-.02,.80 +/-.02,.83 +/-.02, and.80 +/-.02 (C-Index +/- standard error) respectively. Observed mortality for 3,237 isolated CABG patients was 3.7%. For this subgroup, the mean mortality predicted by PS, CL, BA, LR, ST1, and ST2 was 8.4 +/- 7.4, 5.7 +/- 5.9, 6.5 +/- 13.9, 4.5 +/- 6.5, 9.6 +/- 9.1, and 3.0 +/- 3.3 respectively. C-Indexes were.80 +/-.03,.80 +/-.03,.83 +/-.02,.79 +/-.03,.77 +/-.03, and.81 +/-.02 respectively. CONCLUSIONS: Existing CABG models can accurately discriminate outcome about 80 percent of the time. Models developed on a national database and those from non-local databases appear to have validity for our local data set. Predictions can vary widely between models and existing methods for comparing models appear to be inadequate. The methodology presented here is applicable for use with patients undergoing interventions in the cardiac catheterization laboratory.

Journal Article↗
PubMed1997

Modelling of mortality data from a multi-centre study in Japan by means of Poisson regression with error in variables.

BACKGROUND: Death rates of particular categories in epidemiological studies are often based on a small number of occurrences which can be well described by a Poisson distribution. METHOD: We applied this model for the analysis of a multi-centre study in five Japanese counties where the death rates of stomach cancer (ICD-9 code 151) in four age groups are known. In our example some covariates of the cases (e.g. plasma lycopene levels) are unknown values and are estimated from a randomly chosen collective. Therefore these values are subject to a sampling error. The inclusion of errors in variables (e-i-v) into the statistical model can adequately describe such a situation. The model is estimated in a Bayesian framework by means of resampling techniques. RESULTS: Based on the posterior distribution of the parameters the relative risk of stomach cancer is 0.46 (95% confidence interval: 0.23-0.79) comparing the maximum of the population medians of lycopene with the minimum. The estimated overdispersion is close to zero indicating only minor interference with other possible explanatory variables. In addition, we show that inclusion of e-i-v can give more accurate estimates of the parameters even from small sample sizes. CONCLUSIONS: Appropriate statistical methods allow the accurate estimation of relative risks from small sample sizes and from low number of cases. Lycopene plasma levels are good predictors for stomach cancer.

Adult↗
PubMed2005

Protein molecular function prediction by Bayesian phylogenomics.

We present a statistical graphical model to infer specific molecular function for unannotated protein sequences using homology. Based on phylogenomic principles, SIFTER (Statistical Inference of Function Through Evolutionary Relationships) accurately predicts molecular function for members of a protein family given a reconciled phylogeny and available function annotations, even when the data are sparse or noisy. Our method produced specific and consistent molecular function predictions across 100 Pfam families in comparison to the Gene Ontology annotation database, BLAST, GOtcha, and Orthostrapper. We performed a more detailed exploration of functional predictions on the adenosine-5'-monophosphate/adenosine deaminase family and the lactate/malate dehydrogenase family, in the former case comparing the predictions against a gold standard set of published functional characterizations. Given function annotations for 3% of the proteins in the deaminase family, SIFTER achieves 96% accuracy in predicting molecular function for experimentally characterized proteins as reported in the literature. The accuracy of SIFTER on this dataset is a significant improvement over other currently available methods such as BLAST (75%), GeneQuiz (64%), GOtcha (89%), and Orthostrapper (11%). We also experimentally characterized the adenosine deaminase from Plasmodium falciparum, confirming SIFTER's prediction. The results illustrate the predictive power of exploiting a statistical model of function evolution in phylogenomic problems. A software implementation of SIFTER is available from the authors.

Adenosine Deaminase↗
PubMed2003

Population pharmacokinetics and metabolism of midazolam in pediatric intensive care patients.

OBJECTIVE: To determine the pharmacokinetics and metabolism of midazolam in pediatric intensive care patients. DESIGN: Prospective population pharmacokinetic study. SETTING: Pediatric intensive care unit. PATIENTS: Twenty-one pediatric intensive care patients aged between 2 days and 17 yrs. INTERVENTIONS: The pharmacokinetics of midazolam and metabolites were determined during and after a continuous infusion of midazolam (0.05-0.4 mg/kg/hr) for 3.8 hrs to 25 days administered for conscious sedation. MEASUREMENTS AND MAIN RESULTS: Blood samples were taken at different times during and after midazolam infusion for determination of midazolam, 1-OH-midazolam, and 1-OH-midazolam-glucuronide concentrations via high-performance liquid chromatography-ultraviolet detection. A population analysis was conducted via a two-compartment pharmacokinetic model by the NPEM program. The final population model was used to generate individual Bayesian posterior pharmacokinetic parameter estimates. Total body clearance, apparent volume distribution in terminal phase, and plasma elimination half-life were (mean +/- sd, n = 18): 5.0 +/- 3.9 mL/kg/min, 1.7 +/- 1.1 L/kg, and 5.5 +/- 3.5 hrs, respectively. The mean 1-OH-midazolam/midazolam ratio and (1-OH-midazolam + 1-OH-midazolam-glucuronide)/midazolam ratio were 0.14 +/- 0.21 and 1.4 +/- 1.1, respectively. Data from three patients with renal failure, hepatic failure, and concomitant erythromycin-fentanyl therapy were excluded from the final pharmacokinetic analysis. CONCLUSIONS: We describe population and individual midazolam pharmacokinetic parameter estimates in pediatric intensive care patients by using a population modeling approach. Lower midazolam elimination was observed in comparison to other studies in pediatric intensive care patients, probably as a result of differences in study design and patient differences such as age and disease state. Covariates such as renal failure, hepatic failure, and concomitant administration of CYP3A inhibitors are important predictors of altered midazolam and metabolite pharmacokinetics in pediatric intensive care patients. The derived population model can be useful for future dose optimization and Bayesian individualization.

Adolescent↗
PubMed2005

A tractable probabilistic model for Affymetrix probe-level analysis across multiple chips.

MOTIVATION: Affymetrix GeneChip arrays are currently the most widely used microarray technology. Many summarization methods have been developed to provide gene expression levels from Affymetrix probe-level data. Most of the currently popular methods do not provide a measure of uncertainty for the expression level of each gene. The use of probabilistic models can overcome this limitation. A full hierarchical Bayesian approach requires the use of computationally intensive MCMC methods that are impractical for large datasets. An alternative computationally efficient probabilistic model, mgMOS, uses Gamma distributions to model specific and non-specific binding with a latent variable to capture variations in probe affinity. Although promising, the main limitations of this model are that it does not use information from multiple chips and does not account for specific binding to the mismatch (MM) probes. RESULTS: We extend mgMOS to model the binding affinity of probe-pairs across multiple chips and to capture the effect of specific binding to MM probes. The new model, multi-mgMOS, provides improved accuracy, as demonstrated on some bench-mark datasets and a real time-course dataset, and is much more computationally efficient than a competing hierarchical Bayesian approach that requires MCMC sampling. We demonstrate how the probabilistic model can be used to estimate credibility intervals for expression levels and their log-ratios between conditions. AVAILABILITY: Both mgMOS and the new model multi-mgMOS have been implemented in an R package, which is available at http://www.bioinf.man.ac.uk/resources/puma.

Algorithms↗
PubMed1999

Application of Bayesian inference to fMRI data analysis.

The methods of Bayesian statistics are applied to the analysis of fMRI data. Three specific models are examined. The first is the familiar linear model with white Gaussian noise. In this section, the Jeffreys' Rule for noninformative prior distributions is stated and it is shown how the posterior distribution may be used to infer activation in individual pixels. Next, linear time-invariant (LTI) systems are introduced as an example of statistical models with nonlinear parameters. It is shown that the Bayesian approach can lead to quite complex bimodal distributions of the parameters when the specific case of a delta function response with a spatially varying delay is analyzed. Finally, a linear model with auto-regressive noise is discussed as an alternative to that with uncorrelated white Gaussian noise. The analysis isolates those pixels that have significant temporal correlation under the model. It is shown that the number of pixels that have a significantly large auto-regression parameter is dependent on the terms used to account for confounding effects.

Artifacts↗
PubMed2006

Uncertainty in predictions of disease spread and public health responses to bioterrorism and emerging diseases.

Concerns over bioterrorism and emerging diseases have led to the widespread use of epidemic models for evaluating public health strategies. Partly because epidemic models often capture the dynamics of prior epidemics remarkably well, little attention has been paid to how uncertainty in parameter estimates might affect model predictions. To understand such effects, we used Bayesian statistics to rigorously estimate the uncertainty in the parameters of an epidemic model, focusing on smallpox bioterrorism. We then used a vaccination model to translate the uncertainty in the model parameters into uncertainty in which of two vaccination strategies would provide a better response to bioterrorism, mass vaccination, or vaccination of social contacts, so-called "trace vaccination." Our results show that the uncertainty in the model parameters is remarkably high and that this uncertainty has important implications for vaccination strategies. For example, under one plausible scenario, the most likely outcome is that mass vaccination would save approximately 100,000 more lives than trace vaccination. Because of the high uncertainty in the parameters, however, there is also a substantial probability that mass vaccination would save 200,000 or more lives than trace vaccination. In addition to providing the best response to the most likely outcome, mass vaccination thus has the advantage of preventing outcomes that are only slightly less likely but that are substantially more horrific. Rigorous estimates of uncertainty thus can reveal hidden advantages of public health strategies, suggesting that formal uncertainty estimation should play a key role in planning for epidemics.

Bayes Theorem↗
PubMed2003

Robust inference of baseline optical properties of the human head with three-dimensional segmentation from magnetic resonance imaging.

We model the capability of a small (6-optode) time-resolved diffuse optical tomography (DOT) system to infer baseline absorption and reduced scattering coefficients of the tissues of the human head (scalp, skull, and brain). Our heterogeneous three-dimensional diffusion forward model uses tissue geometry from segmented magnetic resonance (MR) data. Handling the inverse problem by use of Bayesian inference and introducing a realistic noise model, we predict coefficient error bars in terms of detected photon number and assumed model error. We demonstrate the large improvement that a MR-segmented model can provide: 2-10% error in brain coefficients (for 2 x 10(6) photons, 5% model error). We sample from the exact posterior and show robustness to numerical model error. This opens up the possibility of simultaneous DOT and MR for quantitative cortically constrained functional neuroimaging.

Bayes Theorem↗
PubMed2003

Bayesian forecasting of oral cyclosporin pharmacokinetics in stable lung transplant recipients with and without cystic fibrosis.

The aims of the current study were (1) to study Neoral pharmacokinetics (PK) in stable lung recipients with or without cystic fibrosis (CF), (2) to compare Neoral PK between these two groups, and (3) to design Bayesian estimators for PK forecasting and dose adjustment in these patients using a limited number of blood samples. The individual PK of 19 adult lung transplant recipients, 9 subjects with CF and 10 subjects without CF, were retrospectively studied. Three profiles obtained within 5 days were available for each patient. A PK model combining a gamma distribution to describe the absorption profile and a two-compartment model were applied. Different exposure indices were estimated using nonlinear regression and Bayesian estimation. The PK model developed reliably described the individual PK of Neoral in lung transplant patients with and without CF, and the values of the first and second half-lives were different in these two populations (lambda(1) = 4.14 +/- 3.01 vs. 2.16 +/- 1.75 h(-1); P < 0.01; lambda(2) = 0.36 +/- 0.11 vs. 0.49 +/- 0.12 h(-1); P < 0.01), while the mean absorption time and standard deviation of absorption time tended to be less in patients with cystic fibrosis (P < 0.1). Also, the patients with CF required higher doses than those without CF to achieve similar drug exposure. Consequently, population modeling was performed in CF and non-CF patients separately. Bayesian estimation allowed accurate prediction of AUC(0-12), AUC(0-4), C(max), and T(max) using three blood samples collected at T0h, T1h, and T3h in both groups. This study demonstrated the applicability and good performance of the PK model previously developed for oral cyclosporin and of the MAP Bayesian estimation of cyclosporin systemic exposure in CF and non-CF patients. Moreover, it is the first to propose a monitoring tool specifically designed for cyclosporin monitoring in patients with CF.

Administration, Oral↗
PubMed2006

Predicting the effect of missense mutations on protein function: analysis with Bayesian networks.

BACKGROUND: A number of methods that use both protein structural and evolutionary information are available to predict the functional consequences of missense mutations. However, many of these methods break down if either one of the two types of data are missing. Furthermore, there is a lack of rigorous assessment of how important the different factors are to prediction. RESULTS: Here we use Bayesian networks to predict whether or not a missense mutation will affect the function of the protein. Bayesian networks provide a concise representation for inferring models from data, and are known to generalise well to new data. More importantly, they can handle the noisy, incomplete and uncertain nature of biological data. Our Bayesian network achieved comparable performance with previous machine learning methods. The predictive performance of learned model structures was no better than a naïve Bayes classifier. However, analysis of the posterior distribution of model structures allows biologically meaningful interpretation of relationships between the input variables. CONCLUSION: The ability of the Bayesian network to make predictions when only structural or evolutionary data was observed allowed us to conclude that structural information is a significantly better predictor of the functional consequences of a missense mutation than evolutionary information, for the dataset used. Analysis of the posterior distribution of model structures revealed that the top three strongest connections with the class node all involved structural nodes. With this in mind, we derived a simplified Bayesian network that used just these three structural descriptors, with comparable performance to that of an all node network.

Algorithms↗
PubMed2006

Inferring global levels of alternative splicing isoforms using a generative model of microarray data.

MOTIVATION: Alternative splicing (AS) is a frequent step in metozoan gene expression whereby the exons of genes are spliced in different combinations to generate multiple isoforms of mature mRNA. AS functions to enrich an organism's proteomic complexity and regulates gene expression. Despite its importance, the mechanisms underlying AS and its regulation are not well understood, especially in the context of global gene expression patterns. We present here an algorithm referred to as the Generative model for the Alternative Splicing Array Platform (GenASAP) that can predict the levels of AS for thousands of exon skipping events using data generated from custom microarrays. GenASAP uses Bayesian learning in an unsupervised probability model to accurately predict AS levels from the microarray data. GenASAP is capable of learning the hybridization profiles of microarray data, while modeling noise processes and missing or aberrant data. GenASAP has been successfully applied to the global discovery and analysis of AS in mammalian cells and tissues. RESULTS: GenASAP was applied to data obtained from a custom microarray designed for the monitoring of 3126 AS events in mouse cells and tissues. The microarray design included probes specific for exon body and junction sequences formed by the splicing of exons. Our results show that GenASAP provides accurate predictions for over one-third of the total events, as verified by independent RT-PCR assays. SUPPLEMENTARY INFORMATION: http://www.psi.toronto.edu/GenASAP.

Algorithms↗
PubMed1980

Prognosis in uveal melanoma with extrascleral extension.

Two thirds of 60 patients followed up after enucleation for uveal melanoma with extrascleral extension eventually died of metastatic disease. Large intraocular tumor size, more malignant cell types, optic nerve invasion, and surgical transection or nonencapsulation of the extraocular tumor margin were found to be significantly correlated with development of metastases. Advanced age at enucleation and large intraocular tumor size were significantly associated with early metastatic death. Recurrence of tumor in the orbit was identified in 10% of the patients and was significantly correlated with large intraocular tumor size and optic nerve invasion. Early exenteration, performed in seven cases, did not improve prognosis. Application of Bayesian methods produced a multifactorial model for prediction of metastatic disease within 13 years after enucleation.

Adolescent↗
PubMed2006

Deterministic and statistical methods for reconstructing multidimensional NMR spectra.

Reconstruction of an image from a set of projections is a well-established science, successfully exploited in X-ray tomography and magnetic resonance imaging. This principle has been adapted to generate multidimensional NMR spectra, with the key difference that, instead of continuous density functions, high-resolution NMR spectra comprise discrete features, relatively sparsely distributed in space. For this reason, a reliable reconstruction can be made from a small number of projections. This speeds the measurements by orders of magnitude compared to the traditional methodology, which explores all evolution space on a Cartesian grid, one step at a time. Speed is of crucial importance for structural investigations of biomolecules such as proteins and for the investigation of time-dependent phenomena. Whereas the recording of a suitable set of projections is a straightforward process, the reconstruction stage can be more problematic. Several practical reconstruction schemes are explored. The deterministic methods-additive back-projection and the lowest-value algorithm-derive the multidimensional spectrum directly from the experimental projections. The statistical search methods include iterative least-squares fitting, maximum entropy, and model-fitting schemes based on Bayesian analysis, particularly the reversible-jump Markov chain Monte Carlo procedure. These competing reconstruction schemes are tested on a set of six projections derived from the three-dimensional 700-MHz HNCO spectrum of a 187-residue protein (HasA) and compared in terms of reliability, absence of artifacts, sensitivity to noise, and speed of computation.

Journal Article↗
PubMed2005

Is anoxic depolarisation associated with an ADC threshold? A Markov chain Monte Carlo analysis.

A Bayesian nonlinear hierarchical random coefficients model was used in a reanalysis of a previously published longitudinal study of the extracellular direct current (DC)-potential and apparent diffusion coefficient (ADC) responses to focal ischaemia. The main purpose was to examine the data for evidence of an ADC threshold for anoxic depolarisation. A Markov chain Monte Carlo simulation approach was adopted. The Metropolis algorithm was used to generate three parallel Markov chains and thus obtain a sampled posterior probability distribution for each of the DC-potential and ADC model parameters, together with a number of derived parameters. The latter were used in a subsequent threshold analysis. The analysis provided no evidence indicating a consistent and reproducible ADC threshold for anoxic depolarisation.

Algorithms↗
PubMed2005

A case study in comparing therapies involving informative drop-out, non-ignorable non-compliance and repeated measurements.

Virtually no comparisons of different psychotherapies with long follow-up times have been carried out until now. The Helsinki Psychotherapy Study is a randomized clinical trial, where patients were monitored for 12 months after the onset of study treatments, of which each lasted approximately 6 months. The patients' psychiatric status was measured at five pre-determined time points during the follow-up period. In general, the analyses of trials are complicated in cases where compliance with the given treatment is incomplete or the drop-out from the follow-up is non-ignorable. In the present study, the quality of the treatment deviated from the protocol for some patients and some patients took auxiliary treatments which had similar effects to the study treatment during the study treatment or follow-up period. This might have resulted in standard intention-to-treat analyses providing excessively conservative or liberal conclusions. Non-compliance may have been non-ignorable in some cases, so subject-specific latent factors may have influenced the outcome both directly and indirectly via compliance behaviour. The most and least healthy patients are the most likely to dropout from the follow-up a priori, so the missing data process is informative. The missing data can partly be augmented with surrogate information collected during interviews with patients who dropped out. A Bayesian hierarchical as-treated model, which uses random-effects-based selection models to account for non-ignorable missing data and non-compliance, was compared with different mixed effects models.

Adult↗
PubMed2006

Effect of CYP2D6 genotype on flecainide pharmacokinetics in Japanese patients with supraventricular tachyarrhythmia.

OBJECTIVE: To examine the effect of CYP2D6 genotype on the pharmacokinetics of flecainide, we conducted a population pharmacokinetic analysis of the data collected during routine therapeutic drug monitoring of Japanese patients with supraventricular tachyarrhythmia. METHODS: Population analysis was performed on retrospective data from 58 patients with normal kidney and liver function treated with oral flecainide for supraventricular tachyarrhythmia. Serum concentrations of flecainide were determined by high-performance liquid chromatography. CYP2D6 genotyping for extensive metabolizer (EM), intermediate metabolizer (IM) and poor metabolizer (PM) alleles was conducted by allele-specific polymerase chain reaction (PCR) and stepdown PCR. WinNonMix was used to estimate oral clearance (CL/F) of flecainide with a one-compartment model for first-order absorption. RESULTS: Body weight, age, sex, serum creatinine concentration (Scr), and CYP2D6 genotype influenced flecainide pharmacokinetics. The CL/F was affected by age (30% reduction in > or =70 years old) and sex (24% reduction in females). The ratios of CL/F for the five CYP2D6 genotypes were: 1.00 (EM/EM), 0.89 (EM/IM), 0.84 (EM/PM), 0.79 (IM/IM), 0.73 (IM/PM). A model including these five covariates reduced the interpatient variability of CL/F from 32.9% (base model) to 17.8%. Using a Bayesian method we estimated that the CL/F in IMs was significantly lower than in homozygous EMs (0.25+/-0.05 l h(-1) kg(-1) vs. 0.37+/-0.08 l h(-1) kg(-1), P<0.05) among male patients under 70 years old. CONCLUSIONS: CYP2D6 genotype, even in IMs, as well as body weight, age, sex, and Scr influence flecainide pharmacokinetics in Japanese patients with supraventricular tachyarrhythmia.

Adult↗
PubMed1996

Non-medical influences on medical decision-making.

BACKGROUND: The influence of non-medical factors on physicians' decision-making has been documented in many observational studies, but rarely in an experimental setting capable of demonstrating cause and effect. We conducted a controlled factorial experiment to assess the influence of non-medical factors on the diagnostic and treatment decisions made by practitioners of internal medicine in two common medical situations. METHOD: One hundred and ninety-two white male internists individually viewed professionally produced video scenarios in which the actor-patient, presenting with either chest pain or dyspnea, possessed various balanced combinations of sex, race, age, socioeconomic status, and health insurance coverage. Physician subjects were randomly drawn from lists of internists in private practice, hospital-based practice, and HMO's, at two levels of experience. RESULTS: The most frequent diagnoses for both chest pain and dyspnea were psychogenic origin and cardiac problems. Smoking cessation was the most frequent treatment recommendation for both conditions. Younger patients (all other factors being the same) were significantly more likely to receive the psychogenic diagnosis. Older patients were more likely to receive the cardiac diagnosis for chest pain, particularly if they were insured. HMO-based physicians were more likely to recommend a follow-up visit for chest pain. Several interactions of patient and physician factors were significant in addition to the main effects. CONCLUSIONS: The variability in decision-making evidenced by physicians in this experiment was not entirely accounted for by strictly rational Bayesian inference (the common prescriptive model for medical decision-making), in-as-much as non-medical factors significantly affected the decisions that they made. There is a need to supplement idealized medical schemata with considerations of social behavior in any comprehensive theory of medical decision-making.

Adult↗