PubMed Health⌕ Search

Biomedical subjects

G Parmigiani

Publications and source records attributed to G Parmigiani.

16 recordsLinked to original sources
PubMed1999

Performing cost-effectiveness analysis by integrating randomized trial data with a comprehensive decision model: application to treatment of acute ischemic stroke.

A recent national panel on cost-effectiveness in health and medicine has recommended that cost-effectiveness analysis (CEA) of randomized controlled trials (RCTs) should reflect the effect of treatments on long-term outcomes. Because the follow-up period of RCTs tends to be relatively short, long-term implications of treatments must be assessed using other sources. We used a comprehensive simulation model of the natural history of stroke to estimate long-term outcomes after a hypothetical RCT of an acute stroke treatment. The RCT generates estimates of short-term quality-adjusted survival and cost and also the pattern of disability at the conclusion of follow-up. The simulation model incorporates the effect of disability on long-term outcomes, thus supporting a comprehensive CEA. Treatments that produce relatively modest improvements in the pattern of outcomes after ischemic stroke are likely to be cost-effective. This conclusion was robust to modifying the assumptions underlying the analysis. More effective treatments in the acute phase immediately following stroke would generate significant public health benefits, even if these treatments have a high price and result in relatively small reductions in disability. Simulation-based modeling can provide the critical link between a treatment's short-term effects and its long-term implications and can thus support comprehensive CEA.

Acute Disease↗
PubMed1999

Is axillary lymph node dissection indicated for early-stage breast cancer? A decision analysis.

PURPOSE: Axillary lymph node dissection (ALND) has been a standard procedure in the management of breast cancer. In a patient with a clinically negative axilla, ALND is performed primarily for staging purposes, to guide adjuvant treatment. Recently, the routine use of ALND has been questioned because the results of the procedure may not change the choice of adjuvant systemic therapy and/or the survival benefit of a change in adjuvant therapy would be small. We constructed a decision model to quantify the benefits of ALND for patients eligible for breast-conserving therapy. METHODS: Patients were grouped by age, tumor size, and estrogen receptor (ER) status. The model uses the Oxford overviews and three combined Cancer and Leukemia Group B studies. We assumed that patients who did not undergo ALND received axillary radiation therapy and that the two procedures are equally effective. All chemotherapy combinations were assumed to be equally efficacious. RESULTS: The largest benefits from ALND are seen in ER-positive women with small primary tumors who might not be candidates for adjuvant chemotherapy if their lymph nodes test negative. Virtually no benefit results in ER-negative women, almost all of whom would receive adjuvant chemotherapy. When adjusted for quality of life (QOL), ALND may have an overall negative impact. In general, the benefits of ALND increase with the expected severity of adjuvant therapy on QOL CONCLUSION: Our model quantifies the benefits of ALND and assists decision making by patients and physicians. The results suggest that the routine use of ALND in breast cancer patients should be reassessed and may not be necessary in many patients.

Adult↗
PubMed1998

Effect of BRCA1 and BRCA2 on the association between breast cancer risk and family history.

BACKGROUND: The discovery of BRCA1 and BRCA2 has led to a reassessment of the association between family history of breast/ovarian cancer and breast cancer risk after controlling for carrier status for mutations in the BRCA1 and BRCA2 genes. We examined whether family history of breast cancer remains a predictive risk factor for this disease after carrier status for BRCA1 and/or BRCA2 mutations is taken into consideration. METHODS: The data are from 4730 case subjects with breast cancer and 4688 control subjects enrolled in the Cancer and Steroid Hormone Study. The probability of being a BRCA1 and/or BRCA2 gene carrier was calculated for each woman. Among predicted noncarriers, logistic regression was used to assess the relationship (odds ratios and 95% confidence intervals [CIs]) between case or control status and family history of breast or ovarian cancer. Estimates of age-specific breast cancer risk are presented by predicted carrier status. RESULTS: Among predicted noncarriers, case subjects were 2.06 times (95% CI = 1.69-2.50) and 1.24 times (95% CI = 1.17-1.32) more likely to report a first-degree or second-degree family history of breast cancer, respectively, than were control subjects. Case subjects were 1.99 times (95% CI = 1.63-2.44), 1.66 times (95% CI = 1.18-2.38), and 2.23 times (95% CI = 0.21-24.65) more likely to report an affected mother, sister, or both, respectively, than were control subjects. A family history of ovarian cancer was not statistically significantly associated with breast cancer risk. Noncarriers were predicted to have a lifetime risk of 9% of developing breast cancer compared with a 63% risk for carriers. CONCLUSIONS: Among women with a moderate family history of breast cancer, i.e., predicted noncarriers of BRCA1 and/or BRCA2 mutations, family history remains a factor in predicting breast cancer risk. In families with breast and ovarian cancers, the aggregation of these two cancers appears to be explained by BRCA1/BRCA2 mutation-carrier probability.

Adult↗
PubMed1998

Protein construct storage: Bayesian variable selection and prediction with mixtures.

Determining optimal conditions for protein storage while maintaining a high level of protein activity is an important question in pharmaceutical research. A designed experiment based on a space-filling design was conducted to understand the effects of factors affecting protein storage and to establish optimal storage conditions. Different model-selection strategies to identify important factors may lead to very different answers about optimal conditions. Uncertainty about which factors are important, or model uncertainty, can be a critical issue in decision-making. We use Bayesian variable selection methods for linear models to identify important variables in the protein storage data, while accounting for model uncertainty. We also use the Bayesian framework to build predictions based on a large family of models, rather than an individual model, and to evaluate the probability that certain candidate storage conditions are optimal.

Bayes Theorem↗
PubMed1998

Determining carrier probabilities for breast cancer-susceptibility genes BRCA1 and BRCA2.

Breast cancer-susceptibility genes BRCA1 and BRCA2 have recently been identified on the human genome. Women who carry a mutation of one of these genes have a greatly increased chance of developing breast and ovarian cancer, and they usually develop the disease at a much younger age, compared with normal individuals. Women can be tested to see whether they are carriers. A woman who undergoes genetic counseling before testing can be told the probabilities that she is a carrier, given her family history. In this paper we develop a model for evaluating the probabilities that a woman is a carrier of a mutation of BRCA1 and BRCA2, on the basis of her family history of breast and ovarian cancer in first- and second-degree relatives. Of special importance are the relationships of the family members with cancer, the ages at onset of the diseases, and the ages of family members who do not have the diseases. This information can be elicited during genetic counseling and prior to genetic testing. The carrier probabilities are obtained from Bayes's rule, by use of family history as the evidence and by use of the mutation prevalences as the prior distribution. In addressing an individual's carrier probabilities, we incorporate uncertainty about some of the key inputs of the model, such as the age-specific incidence of diseases and the overall prevalence of mutations. There is some evidence that other, undiscovered genes may be important in explaining familial breast cancer. Users of the current version of the model should be aware of this limitation. The methodology that we describe can be extended to more than two genes, should data become available about other genes.

BRCA2 Protein↗
PubMed1998

Missense mutations in disease genes: a Bayesian approach to evaluate causality.

The problem of interpreting missense mutations of disease-causing genes is an increasingly important one. Because these point mutations result in alteration of only a single amino acid of the protein product, it is often unclear whether this change alone is sufficient to cause disease. We propose a Bayesian approach that utilizes genetic information on affected relatives in families ascertained through known missense-mutation carriers. This method is useful in evaluating known disease genes for common disease phenotypes, such as breast cancer or colorectal cancer. The posterior probability that a missense mutation is disease causing is conditioned on the relationship of the relatives to the proband, the population frequency of the mutation, and the phenocopy rate of the disease. The approach is demonstrated in two cancer data sets: BRCA1 R841W and APC I1307K. In both examples, this method helps establish that these mutations are likely to be disease causing, with Bayes factors in favor of causality of 5.09 and 66.97, respectively, and posterior probabilities of .836 and .985. We also develop a simple approximation for rare alleles and consider the case of unknown penetrance and allele frequency.

Adenomatous Polyposis Coli↗
PubMed1997

The Stroke Prevention Policy Model: linking evidence and clinical decisions.

Simulation models that support decision and cost-effectiveness analysis can further the goals of evidence-based medicine by facilitating the synthesis of information from several sources into a single comprehensive structure. The Stroke Prevention Policy Model (SPPM) performs this function for the clinical and policy questions that surround stroke prevention. This paper first describes the basic structure and functions of the SPPM, concentrating on the role of large databases (broadly defined as any database that contains many patients, regardless of study design) in providing the SPPM inputs. Next, recognizing that the use of modeling continues to be a source of some controversy in the medical community, it discusses the philosophical underpinnings of the SPPM. Finally, it discusses conclusions in the context of both stroke prevention and other complex medical decisions. We conclude that despite the difficulties in developing comprehensive models (for example, the length and complexity of model development and validation processes, the proprietary nature of data sources, and the necessity for developing new software), the benefits of such models exceed the costs of continuing to rely on more conventional methods. Although they should not replace the clinician in decision making, comprehensive models based on the best available evidence from large databases can support decision making in medicine.

Cerebrovascular Disorders↗
PubMed1997

Probability of carrying a mutation of breast-ovarian cancer gene BRCA1 based on family history.

BACKGROUND: Heritable mutations of the breast cancer gene BRCA1 are rare, occurring in fewer than 1% of women in the general population, and therefore account for a small proportion of cases of breast and ovarian cancers. Nevertheless, the presence of such mutations is highly predictive of the development of these cancers. PURPOSE: We developed and applied a mathematic model for calculating the probability that a woman with a family history of breast and/or ovarian cancer carries a mutation of BRCA1. METHODS AND RESULTS: As a basis for the model, we use Mendelian genetics and apply Bayes' theorem to information on the family history of these diseases. Of importance are the exact relationships of all family members, including both affected and unaffected members, and ages at diagnosis of the affected members and current ages of the unaffected members. We used available estimates of BRCA1 mutation frequencies in the general population and age-specific incidence rates of breast and ovarian cancers in carriers and noncarriers of mutations to estimate the probability that a particular member of the family carries a mutation. This probability is based on cancer statuses of all first- and second-degree relatives. We first describe the model by considering single individuals: a woman diagnosed with breast and/or ovarian cancer and also a woman free of cancer. We next considered two artificial and two actual family histories and addressed the sensitivity of our calculations to various assumptions. Particular relationships of family members with and without cancer can have a substantial impact on the probability of carrying a susceptibility gene. Ages at diagnosis of affected family members and their types of cancer are also important. A woman with two primary cancers can have a probability of carrying a mutation in excess of 80%, even with no other information about family history. The number and relationships of unaffected members, along with their current ages or ages at death, are critical determinants of one's carrier probability. An affected woman with several cancers in her family can have a probability of carrying a mutation that ranges from close to 100% to less than 5%. CONCLUSION: Our model gives informative and specific probabilities that a particular woman carries a mutation. IMPLICATIONS: This model focuses on mutations in BRCA1 and assumes that all other breast cancer is sporadic. With the cloning of BRCA2, we now know that this assumption is incorrect. We have adjusted the model to include BRCA2, but the use of this version must await publication of penetrance data for BRCA2, including those for male breast cancer that are apparently associated with BRCA2 but not with BRCA1. The current model is, nevertheless, appropriate and useful. Of principal importance is its potential and that of improved versions for aiding women and their health care providers in assessing the need for genetic testing.

Adult↗
PubMed1997

The clinical significance of positive blood cultures in the 1990s: a prospective comprehensive evaluation of the microbiology, epidemiology, and outcome of bacteremia and fungemia in adults.

To assess changes since the mid-1970s, we reviewed 843 episodes of positive blood cultures in 707 patients with septicemia. The five most common pathogens were Staphylococcus aureus, Escherichia coli, coagulase-negative staphylococci (CNS), Klebsiella pneumoniae, and Enterococcus species. Although CNS were isolated most often, only 12.4% were clinically significant. Half of all episodes were nosocomial, and a quarter had no recognized source. Leading identifiable sources included intravenous catheters, the respiratory and genitourinary tracts, and intraabdominal foci. Septicemia-associated mortality was 17.5%. Patients who received appropriate antimicrobial therapy throughout the course of infection had the lowest mortality (13.3%). Multivariate analysis showed that age (relative risk [RR], 1.80), microorganism (RR, 2.27), source of infection (RR, 2.86), predisposing factors (RR, 1.98), blood pressure (RR, 2.29), body temperature (RR, 2.04), and therapy (RR, 2.72) independently influenced outcome. Bloodstream infections in the 1990s are notable for the increased importance of CNS as both contaminants and pathogens, the proportionate increase in fungi and decrease in anaerobes as pathogens, the emergence of Mycobacterium avium complex as an important cause of bacteremia in patients with advanced human immunodeficiency virus infection, and the reduction in mortality associated with infection.

Adolescent↗
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↗