AIMSWe determined the feasibility and diagnostic performance of segmental 18F-flurpiridaz myocardial blood flow (MBF) measurement by positron emission tomography (PET) compared with the standard territory method, and assessed whether flow metrics provide incremental diagnostic value beyond relative perfusion quantitation (PQ).METHODS AND RESULTSAll evaluable pharmacological stress patients from the Phase III trial of 18F-flurpiridaz were included (n = 245) and blinded flow metrics obtained. For each coronary territory, the segmental flow metric was defined as the lowest 17-segment stress MBF (SMBF), myocardial flow reserve (MFR), or relative flow reserve (RFR) value. Diagnostic performances of segmental and territory MBF metrics were compared by receiver operating characteristic (ROC) areas under the curve (AUC). A multiple logistic model was used to evaluate whether flow metrics provided incremental diagnostic value beyond PQ alone. The diagnostic performances of segmental flow metrics were higher than their territory counterparts; SMBF AUC = 0.761 vs. 0.737; MFR AUC = 0.699 vs. 0.676; and RFR AUC = 0.716 vs. 0.635, respectively (P < 0.001 for all). Similar results were obtained for per-vessel coronary artery disease (CAD) ≥70% stenosis categorization and per-patient analyses. Combinatorial analyses revealed that only SMBF significantly improved the diagnostic performance of PQ in CAD ≥50% stenoses, with PQ AUC = 0.730, PQ + segmental SMBF AUC = 0.782 (P < 0.01), and PQ + territory SMBF AUC = 0.771 (P < 0.05). No flow metric improved diagnostic performance when combined with PQ in CAD ≥70% stenoses.CONCLUSIONAssessment of segmental MBF metrics with 18F-flurpiridaz is feasible and improves flow-based epicardial CAD detection. When combined with PQ, only SMBF provides additive diagnostic performance in moderate CAD.

652 Objectives: Myocardial blood flow (MBF) quantitation by positron emission tomography (PET) provides an integrated assessment of epicardial and microvascular coronary artery disease (CAD). MBF quantitation with currently available PET radiopharmaceuticals is a powerful prognostic indicator, with potential utility in increasing diagnostic sensitivity for the detection of CAD and identification of 3-vessel CAD. We previously developed and validated a single-compartment model of MBF quantitation with flurpiridaz in humans over a wide range of flow conditions (J Nucl Med 2014;55:1438-44). In this study, we sought to develop an automated approach of flurpiridaz MBF quantitation, and to evaluate its diagnostic performance in a subgroup of 245 patients from a phase III multi-center flurpiridaz clinical trial. Methods: All patients from the phase III 301 clinical trial of flurpiridaz PET MPI who underwent rest-pharmacological stress and whose studies were suitable for flow quantitation were evaluated (n=245). Sixteen patients (8 males) were selected as reference normals (RN) based on having < 35% stenosis in any vessel, left ventricular ejection fraction ≥ 50%, and no history of hypertension, diabetes mellitus, myocardial infarction, coronary revascularization, or heart failure. The remaining n=229 were used for diagnostic assessment. The reference standard was blinded quantitative invasive coronary angiography. Results: To assess global flow, and flow in the combined (pooled) individual coronary territories, patients were subgrouped according to the maximal % narrowing of a given coronary artery as RN (n=16 and n=48), 0-29% (n=65 and n=371), 30-49% (n=56 and n=153), 50-69% (n=42 and n=77), and 70-100% (n=66 and n=86, respectively). Global stress MBF decreased significantly (P<0.001) in subgroups with increasing % stenosis; RN (2.22 ± 0.60), 0-29% (2.02 ± 0.56), 30-49% (1.90 ± 0.57), 50-69% (1.74 ± 0.52), and 70-100% (1.49 ± 0.46). Furthermore, global MFR (stress MBF / rest MBF) had a similar significant (P<0.001) trend; RN (3.10 ± 0.68), 0-29% (2.95 ± 0.97), 30-49% (2.90 ± 0.85), 50-69% (2.87 ± 1.08), and 70-100% (2.42 ± 0.84). An analogous significant (P<0.001) pattern was also observed for the pooled territories, with stress MBF’s; RN (2.22 ± 0.60), 0-29% (1.96 ± 0.59), 30-49% (1.74 ± 0.54), 50-69% (1.55 ± 0.50), and 70-100% (1.31 ± 0.50), and with MFR’s; RN (3.10 ± 0.68), 0-29% (2.93 ± 0.99), 30-49% (2.81 ± 0.89), 50-69% (2.64 ± 0.96), and 70-100% (2.11 ± 0.83). We further determined receiver operating characteristic (ROC) areas under the curve (AUC) of stress MBF and MFR. In global flow, the stress MBF AUC was significantly greater than the MFR AUC (0.70 vs. 0.63, P=0.007) defining CAD as ≥ 50%, with similar results for CAD ≥ 70% (0.72 vs. 0.66, P=0.033). In pooled territories, the stress MBF AUC was significantly greater than the MFR AUC (0.73 vs. 0.67, P=0.002) defining CAD as ≥ 50%, but not for CAD ≥ 70% (0.75 vs. 0.74, P=0.31). Using CAD ≥ 50% and ≥ 70% as the reference, the optimal cutoff points for the highest sensitivity and specificity were 1.8 and 1.7 for global stress MBF, 2.8 and 2.7 for global MFR, 1.7 and 1.6 for pooled territories stress MBF, and 2.6 and 2.5 for pooled territories MFR, respectively. Conclusions: We have developed an automated method for quantitation of global and regional myocardial blood flow (MBF) by flurpiridaz PET MPI which demonstrates discrimination between subgroups with different degrees of angiographic stenosis. The diagnostic accuracy of stress MBF was superior to myocardial flow reserve (MFR) when assessing global flow with CAD ≥ 50% and ≥ 70% as the reference, whereas in pooled territories, stress MBF was superior to MFR only in CAD ≥ 50%. Finally, we established reference values for determination of significant CAD both for stress MBF and for MFR.

Coastal zones are popular recreational areas that substantially contribute to social welfare. Managers can use information about specific environmental features that people value, and how these might change under different management scenarios, to spatially target actions to areas of high current or potential value. We explored how snorkelers' experience would be affected by separate and combined land and marine management actions in West Maui, Hawaiʻi, using a Bayesian belief network (BBN) and a spatially explicit ecosystem services model. The BBN simulates the attractiveness of a site for recreation by combining snorkeler preferences for coastal features with expert opinions on ecological dynamics, snorkeler behavior, and management actions. A choice experiment with snorkelers elucidated their preferences for sites with better ecological and water-quality conditions. Linking the economic elicitation to the spatially explicit BBN to evaluate land-sea management scenarios provides specific guidance on where and how to act in West Maui to maximize ecosystem service returns. Improving coastal water quality through sediment runoff and cesspool effluent reductions (land management), and enhancing coral reef ecosystem conditions (marine management) positively affected overall snorkeling attractiveness across the study area, but with differential results at specific sites. The highest improvements were attained through joint land-sea management, driven by strong efforts to increase fish abundance and reduce sediment; however, the effects of management at individual beaches varied.

Background Computerized methodologies standardize the myocardial perfusion imaging (MPI) interpretation process. Methods To develop an automated relative perfusion quantitation approach for 18 F-flurpiridaz, PET MPI studies from all phase III trial participants of 18 F-flurpiridaz were divided into 3 groups. Count distributions were obtained in N = 40 normal patients undergoing pharmacological or exercise stress. Then, N = 90 additional studies were selected in a derivation group. Following receiver operating characteristic curve analysis, various standard deviations below the mean normal were used as cutoffs for significant CAD, and interobserver variability determined. Finally, diagnostic performance was compared between blinded visual readers and blinded derivations of automated relative quantitation in the remaining N = 548 validation patients. Results Both approaches yielded comparable accuracies for the detection of global CAD, reaching 71% and 72% by visual reads, and 72% and 68% by automated relative quantitation, when using CAD ≥ 70% or ≥ 50% stenosis for significance, respectively. Similar results were observed when analyzing individual coronary territories. In both pharmacological and exercise stress, automated relative quantitation demonstrated significantly more interobserver agreement than visual reads. Conclusions Our automated method of 18 F-flurpiridaz relative perfusion analysis provides a quantitative, objective, and highly reproducible assessment of PET MPI in normal and CAD subjects undergoing either pharmacological or exercise stress.

102 Objectives: Computerized methodologies help standardize the myocardial perfusion imaging (MPI) interpretation process, known to be subject to reader variability. We sought to adapt our previously developed automated relative perfusion quantitation method to 18F-Flurpiridaz MPI, a novel positron emission tomography (PET) radiopharmaceutical, determine normal limits and criteria for abnormality, and compare performance metrics of automation with that of expert visual reads. Methods: PET MPI studies from all n=678 evaluable participants of the phase 301 trial of 18F-Flurpiridaz were included. Following adaptation of an automated method, normal limits for automated relative quantitation were obtained in n=40 patients undergoing pharmacological or exercise stress. The polar map approach was used to represent the left ventricular myocardial 18F-Flurpiridaz perfusion distribution. Then, n=90 additional studies were selected in a derivation group to analyze both normal and coronary artery disease (CAD) patients categorized according to ≥ 70% and ≥ 50% stenosis on invasive coronary angiography. Following receiver operating characteristic curve analysis, various standard deviations below the mean normal for all pixels in the myocardium were used as cutoffs for significant CAD, and interobserver variability determined. Accuracy, sensitivity, and specificity, was compared between blinded visual readers from the trial core and blinded derivations of automated relative quantitation in all the remaining n=548 validation patients. Results: Overall, both approaches yielded comparable accuracies for the detection of global CAD, reaching 71% and 72% by visual reads, and 72% and 68% by automated relative quantitation, when using CAD ≥ 70% or ≥ 50% stenosis for significance, respectively. Similar results were observed when analyzing individual coronary territories. In both forms of stress, automated relative quantitation demonstrated significantly more interobserver agreement than visual reads. Conclusions: We present a comprehensive automated method of relative perfusion analysis, providing a quantitative, objective, and highly reproducible, assessment of 18F-Flurpiridaz PET MPI, paving the way for future clinical applicability.

Foundations of Stated Preference Elicitation: Consumer Behavior and Choice-based Conjoint Analysis

We describe welfare calculations when an environmental injury reduces trips to undamaged sites as well as those that were damaged. The welfare loss is (1) underestimated when standard welfare formulas are applied only to damaged sites but (2) overestimated when these formulas are applied to all sites with lost trips. We provide a formula that appropriately accounts for the lost trips to undamaged sites. Differences among the procedures are illustrated through hypothetical scenarios that differ in lost trips to undamaged sites. We apply the method under linear demand to aggregate estimates of shoreline-use losses from the Deepwater Horizon oil spill. (JEL D61, D81)

This paper examines sources of correlation among utility coefficients in models allowing for random heterogeneity, including correlation that is induced by random scale heterogeneity. We distinguish the capabilities and limitations of various models, including mixed logit, generalized multinomial logit (G-MNL), latent class, and scale-adjusted latent class. We demonstrate that (i) mixed logit allows for all forms of correlation, including scale heterogeneity, (ii) G-MNL is a restricted form of mixed logit that, with an appropriate implementation, can allow for scale heterogeneity but (in its typical form) not other sources of correlation, (iii) none of the models disentangles scale heterogeneity from other sources of correlation, and (iv) models that assume that the only source of correlation is scale heterogeneity necessarily capture, in the estimated scale parameter, whatever other sources of correlation exist.

This paper examines sources of correlation among utility coefficients in models allowing for random heterogeneity, including correlation that is induced by random scale heterogeneity. We distinguish the capabilities and limitations of various models, including mixed logit, generalized multinomial logit (G-MNL), latent class, and scale-adjusted latent class. We demonstrate that (i) mixed logit allows for all forms of correlation, including scale heterogeneity, (ii) G-MNL is a restricted form of mixed logit that allows for scale heterogeneity but (in its typical form) not other sources of correlation, (iii) none of the models disentangles scale heterogeneity from other sources of correlation, and (iv) models that assume that the only source of correlation is scale heterogeneity necessarily capture, in the estimated scale parameter, whatever other sources of correlation exist.

AbstractPatents without established market values (e. g., no negotiated royalty rates) are often valued by comparing the number of citations the patent has received to the numbers received by other patents whose market values are established. For recently-issued patents, which have not had time to accumulate citations, this procedure can be noisy or even inapplicable. The current paper generalizes this valuation method to incorporate patent characteristics that relate to the number of citations the patent is expected to obtain in the future. We estimate statistical models in which the explanatory variables are observable characteristics of the patent at a given time, and the dependent variable is the number of citations that the patent receives after that date. Using several examples, we demonstrate a procedure for patent valuation that incorporates the statistical results, such that the valuation reflects the number of citations the patent has already received as well as the number it is expected, based on its characteristics, to receive in the future.

This article extends the discussion introduced in this journal by Haab et al. (2013), who critique a 2012 article on contingent valuation (CV) by Hausman. We agree with Haab et al. "that the intellectual debate over [CV] is ongoing and that plenty of work remains to be done for the truly curious researcher" (593). However, our paper illuminates the limitations of current approaches to guide future research.

This paper presents a flexible procedure for representing the distribution of random parameters in mixed logit models. A logit formula is specified for the mixing distribution, in addition to its use for the choice probabilities. The properties of logit assure positivity and provide the normalizing constant for the mixing distribution. Any mixing distribution can be approximated to any degree of accuracy by this specification. The researcher defines variables to describe the shape of the mixing distribution, using flexible forms such as polynomials, splines, and step functions. The gradient of the log-likelihood is easy to calculate, which facilitates estimation. The procedure is illustrated with data on consumers' choice among video streaming services.

A review of 116 travel cost models finds that, of the studies that report their practice, about half include depreciation in their calculation of vehicle costs and half do not, with none giving a justification for either approach. We examine empirically whether depreciation is related to households' decisions of how much to drive. Using a sample of over 200,000 US households, we find that, relative to fuel costs, depreciation has a small effect on the amount that households drive. This finding is consistent with households' considering depreciation as primarily a fixed rather than marginal cost.

A review of 116 travel cost models finds that, of the studies that report their practice, about half include depreciation in their calculation of vehicle costs and half do not, with none giving a justification for either approach. We examine empirically whether depreciation is related to households' decisions of how much to drive. Using a sample of over 200,000 US households, we find that, relative to fuel costs, depreciation has a small effect on the amount that households drive. This finding is consistent with households' considering depreciation as primarily a fixed rather than marginal cost.

The attributes that consumers anticipate or expect when choosing among alternatives (i.e., prior to consumption) often differ from the attributes that they actually experience when consuming their chosen alternative. This paper describes, and illustrates with several examples, the calculation of consumer surplus in this situation and the loss in consumer surplus due to the imperfect foreknowledge about attributes. The procedures are useful in many settings, such as the assessment of damages for false advertising and the analysis of informational policies.

For a contingent valuation (CV) study of a bundle of water quality improvements, this paper tests whether the sum of the estimated willingness to pay (WTP) for each individual part of the package, evaluated incrementally, equals the estimated WTP for the entire bundle-as implied by standard utility theory. This is the first application of the adding-up test using incremental parts and Carson and Groves's (2007) procedures to induce truthful respondents' answers. The test is failed, which implies that either the CV method did not elicit truthful answers or that respondents' true preferences are inconsistent with standard utility theory. (JEL Q51)

This paper re-examines the discrete choice methods and stated preference elicitation procedures that are commonly used in choice-based conjoint analysis. The aim is to clarify their domains of applicability and provide reliable techniques for data collection and analysis. CONTENTS Topic Page Preface 2 1 Choice Behavior and Consumer Welfare 2 1.1 Utility 5 1.2 Choice Probabilities 8 1.3 Consumer Welfare 10 1.4 Estimation 12 2 Some History of Stated Preference Elicitation 13 3 Choice-Based Conjoint Analysis (CBC) 15 3.1. Sampling and Recruitment 18 3.2. Design of Conjoint Experiments 19 3.3. Subject Training 21 3.4. Incentive Alignment 22 3.5. Reconciliation and Validation 23 3.6. A Conjoint Study Example 23 3.7. Stated Perceptions, Expectations, and Well-Being 25 4 Maximum Simulated Likelihood (MSL) Analysis of CBC Data 26 5 Hierarchical Bayes (HB) Estimation 29 5.1. Bayesian Estimation 30 5.2. Selecting Priors and Computing Posteriors 36 5.3. Hierarchical Bayes Methods 43 5.4. Policy Simulations 46 5.5. A Monte Carlo Example 47 6 An Empirical CBC Study using MSL and HB Methods 51 7 Caveats and Conclusions 55