Prediction markets supporting technology assessment

In this thesis, we study the use of prediction markets for technology assessment. We particularly focus on their ability to assess complex issues, the design constraints required for such applications and their efficacy compared to traditional techniques. To achieve this, we followed a design science research paradigm, iteratively developing, instantiating, evaluating and refining the design of our artifacts. This allowed us to make multiple contributions, both practical and theoretical. We first showed that prediction markets are adequate for properly assessing complex issues. We also developed a typology of design factors and design propositions for using these markets in a technology assessment context. Then, we showed that they are able to solve some issues related to the R&D portfolio management process and we proposed a roadmap for their implementation. Finally, by comparing the instantiation and the results of a multi-criteria decision method and a prediction market, we showed that the latter are more efficient, while offering similar results. We also proposed a framework for comparing forecasting methods, to identify the constraints based on contingency factors. In conclusion, our research opens a new field of application of prediction markets and should help hasten their adoption by enterprises. Résumé français: Dans cette thèse, nous étudions l'utilisation de marchés de prédictions pour l'évaluation de nouvelles technologies. Nous nous intéressons plus particulièrement aux capacités des marchés de prédictions à évaluer des problématiques complexes, aux contraintes de conception pour une telle utilisation et à leur efficacité par rapport à des techniques traditionnelles. Pour ce faire, nous avons suivi une approche Design Science, développant itérativement plusieurs prototypes, les instanciant, puis les évaluant avant d'en raffiner la conception. Ceci nous a permis de faire de multiples contributions tant pratiques que théoriques. Nous avons tout d'abord montré que les marchés de prédictions étaient adaptés pour correctement apprécier des problématiques complexes. Nous avons également développé une typologie de facteurs de conception ainsi que des propositions de conception pour l'utilisation de ces marchés dans des contextes d'évaluation technologique. Ensuite, nous avons montré que ces marchés pouvaient résoudre une partie des problèmes liés à la gestion des portes-feuille de projets de recherche et développement et proposons une feuille de route pour leur mise en oeuvre. Finalement, en comparant la mise en oeuvre et les résultats d'une méthode de décision multi-critère et d'un marché de prédiction, nous avons montré que ces derniers étaient plus efficaces, tout en offrant des résultats semblables. Nous proposons également un cadre de comparaison des méthodes d'évaluation technologiques, permettant de cerner au mieux les besoins en fonction de facteurs de contingence. En conclusion, notre recherche ouvre un nouveau champ d'application des marchés de prédiction et devrait permettre d'accélérer leur adoption par les entreprises.

Prediction of Infant Drug Exposure Through Breastfeeding: Population PK Modeling and Simulation of Fluoxetine Exposure.

The likelihood of significant exposure to drugs in infants through breast milk is poorly defined, given the difficulties of conducting pharmacokinetics (PK) studies. Using fluoxetine (FX) as an example, we conducted a proof-of-principle study applying population PK (popPK) modeling and simulation to estimate drug exposure in infants through breast milk. We simulated data for 1,000 mother-infant pairs, assuming conservatively that the FX clearance in an infant is 20% of the allometrically adjusted value in adults. The model-generated estimate of the milk-to-plasma ratio for FX (mean: 0.59) was consistent with those reported in other studies. The median infant-to-mother ratio of FX steady-state plasma concentrations predicted by the simulation was 8.5%. Although the disposition of the active metabolite, norfluoxetine, could not be modeled, popPK-informed simulation may be valid for other drugs, particularly those without active metabolites, thereby providing a practical alternative to conventional PK studies for exposure risk assessment in this population.

Validation of a clinical prediction score for ruling out coronary heart disease in primary care patients with chest pain.

Background: A patient's chest pain raises concern for the possibility of coronary heart disease (CHD). An easy to use clinical prediction rule has been derived from the TOPIC study in Lausanne. Our objective is to validate this clinical score for ruling out CHD in primary care patients with chest pain. Methods: This secondary analysis used data collected from a oneyear follow-up cohort study attending 76 GPs in Germany. Patients attending their GP with chest pain were questioned on their age, gender, duration of chest pain (1-60 min), sternal pain location, pain increases with exertion, absence of tenderness point at palpation, cardiovascular risks factors, and personal history of cardiovascular disease. Area under the curve (ROC), sensitivity and specificity of the Lausanne CHD score were calculated for patients with full data. Results: 1190 patients were included. Full data was available for 509 patients (42.8%). Missing data was not related to having CHD (p = 0.397) or having a cardiovascular risk factor (p = 0.275). 76 (14.9%) were diagnosed with a CHD. Prevalence of CHD were respectively of 68/344 (19.8%), 2/62 (3.2%), 6/103 (5.8%) in the high, intermediate and low risk category. ROC was of 72.9 (CI95% 66.8; 78.9). Ruling out patients with low risk has a sensitivity of 92.1% (CI95% 83.0; 96.7) and a specificity of 22.4% (CI95% 18.6%; 26.7%). Conclusion: The Lausanne CHD score shows reasonably good sensitivity and can be used to rule out coronary events in patients with chest pain. Patients at risk of CHD for other rarer reasons should nevertheless also be investigated.

Ruling out coronary artery disease in primary care: development and validation of a simple prediction rule.

BACKGROUND: Chest pain can be caused by various conditions, with life-threatening cardiac disease being of greatest concern. Prediction scores to rule out coronary artery disease have been developed for use in emergency settings. We developed and validated a simple prediction rule for use in primary care. METHODS: We conducted a cross-sectional diagnostic study in 74 primary care practices in Germany. Primary care physicians recruited all consecutive patients who presented with chest pain (n = 1249) and recorded symptoms and findings for each patient (derivation cohort). An independent expert panel reviewed follow-up data obtained at six weeks and six months on symptoms, investigations, hospital admissions and medications to determine the presence or absence of coronary artery disease. Adjusted odds ratios of relevant variables were used to develop a prediction rule. We calculated measures of diagnostic accuracy for different cut-off values for the prediction scores using data derived from another prospective primary care study (validation cohort). RESULTS: The prediction rule contained five determinants (age/sex, known vascular disease, patient assumes pain is of cardiac origin, pain is worse during exercise, and pain is not reproducible by palpation), with the score ranging from 0 to 5 points. The area under the curve (receiver operating characteristic curve) was 0.87 (95% confidence interval [CI] 0.83-0.91) for the derivation cohort and 0.90 (95% CI 0.87-0.93) for the validation cohort. The best overall discrimination was with a cut-off value of 3 (positive result 3-5 points; negative result <or= 2 points), which had a sensitivity of 87.1% (95% CI 79.9%-94.2%) and a specificity of 80.8% (77.6%-83.9%). INTERPRETATION: The prediction rule for coronary artery disease in primary care proved to be robust in the validation cohort. It can help to rule out coronary artery disease in patients presenting with chest pain in primary care.

HIV disclosure and nondisclosure among migrant women from sub-Saharan Africa living in Switzerland.

No study to date has focused specifically on the reasons for and against disclosure of HIV-positive status among sub-Saharan migrant women. Thirty HIV-positive women from 11 sub-Saharan countries living in French-speaking Switzerland participated in semi-structured individual interviews. The reasons women reported for disclosure or nondisclosure of their HIV serostatus were classified into three categories: social, medical, and ethical. The women identified the stigma associated with HIV as a major social reason for nondisclosure. However, this study identifies new trends related to disclosure for medical and ethical reasons. Being undetectable played an important role in the life of sub-Saharan migrant women, and analysis revealed their medical reasons for both disclosure and nondisclosure. Disclosure to new sexual partners occurred when women had a more positive perception about HIV and when they believed themselves to be in a long-term relationship. Women reported nondisclosure to family members when they did not need help outside the support provided by the medical and social fields. The results on ethical reasons suggested that challenging stigma was a reason for disclosure. Since the women' perceptions on HIV changed when they came to see it as a chronic disease, disclosure occurred in an attempt to normalize life with HIV in their communities in migration and to challenge racism and discrimination. Our findings can help health providers better understand the communication needs of sub-Saharan migrant women with respect to HIV/AIDS and sexuality and offer them adequate disclosure advice that takes into account migration and gender issues.

Outcome Prediction after Cardiac Arrest Treated with Hypothermia

Rationale: Clinical and electrophysiological prognostic markers of brain anoxia have been mostly evaluated in comatose survivors of out hospital cardiac arrest (OHCA) after standard resuscitation, but their predictive value in patients treated with mild induced hypothermia (IH) is unknown. The objective of this study was to identify a predictive score of independent clinical and electrophysiological variables in comatose OHCA survivors treated with IH, aiming at a maximal positive predictive value (PPV) and a high negative predictive value (NPV) for mortality. Methods: We prospectively studied consecutive adult comatose OHCA survivors from April 2006 to May 2009, treated with mild IH to 33-34_C for 24h at the intensive care unit of the Lausanne University Hospital, Switzerland. IH was applied using an external cooling method. As soon as subjects passively rewarmed (body temperature >35_C) they underwent EEG and SSEP recordings (off sedation), and were examined by experienced neurologists at least twice. Patients with status epilepticus were treated with AED for at least 24h. A multivariable logistic regression was performed to identify independent predictors of mortality at hospital discharge. These were used to formulate a predictive score. Results: 100 patients were studied; 61 died. Age, gender and OHCA etiology (cardiac vs. non-cardiac) did not differ among survivors and nonsurvivors. Cardiac arrest type (non-ventricular fibrillation vs. ventricular fibrillation), time to return of spontaneous circulation (ROSC) >25min, failure to recover all brainstem reflexes, extensor or no motor response to pain, myoclonus, presence of epileptiform discharges on EEG, EEG background unreactive to pain, and bilaterally absent N20 on SSEP, were all significantly associated with mortality. Absent N20 was the only variable showing no false positive results. Multivariable logistic regression identified four independent predictors (Table). These were used to construct the score, and its predictive values were calculated after a cut-off of 0-1 vs. 2-4 predictors. We found a PPV of 1.00 (95% CI: 0.93-1.00), a NPV of 0.81 (95% CI: 0.67-0.91) and an accuracy of 0.93 for mortality. Among 9 patients who were predicted to survive by the score but eventually died, only 1 had absent N20. Conclusions: Pending validation in a larger cohort, this simple score represents a promising tool to identify patients who will survive, and most subjects who will not, after OHCA and IH. Furthermore, while SSEP are 100% predictive of poor outcome but not available in most hospitals, this study identifies EEG background reactivity as an important predictor after OHCA. The score appears robust even without SSEP, suggesting that SSEP and other investigations (e.g., mismatch negativity, serum NSE) might be principally needed to enhance prognostication in the small subgroup of patients failing to improve despite a favorable score.

Seasonal changes in 24-h and basal energy expenditures in rural Gambian men as measured in a respiration chamber.

Adaptation of 24-h energy expenditure (24-h EE) to seasonal variations in food availability was studied, by using a respiration chamber, in 18 rural Gambian men on three occasions: period 1--at the end of the rainy season, which is characterized by low food availability; period 2--during the nutritionally favorable dry season; and period 3--at the onset of the following rainy season. From periods 1 to 2 body weight increased by 2.8 +/- 0.4 kg, and a rise in 24-h EE was observed (from 8556 +/- 212 kJ/d to 9166 +/- 224 kJ/d), which was correlated to weight change (r = 0.73, P less than 0.001). During period 3, 24-h EE averaged 8740 +/- 194 kJ/d. Diet-induced thermogenesis increased significantly from periods 1 to 2 (5.9 +/- 0.5% to 8.2 +/- 0.8%) and subsequently decreased to 3.6 +/- 0.6% during period 3. In rural Gambian men, metabolic adaptations in response to seasonal changes in food availability are reflected by a decrease in body weight, mainly manifested by a loss of fat-free mass accompanied by a decreased 24-h EE and a lowered diet-induced thermogenesis.

Rgtsp: a generalized top scoring pairs package for class prediction.

SUMMARY: A top scoring pair (TSP) classifier consists of a pair of variables whose relative ordering can be used for accurately predicting the class label of a sample. This classification rule has the advantage of being easily interpretable and more robust against technical variations in data, as those due to different microarray platforms. Here we describe a parallel implementation of this classifier which significantly reduces the training time, and a number of extensions, including a multi-class approach, which has the potential of improving the classification performance. AVAILABILITY AND IMPLEMENTATION: Full C++ source code and R package Rgtsp are freely available from http://lausanne.isb-sib.ch/~vpopovic/research/. The implementation relies on existing OpenMP libraries.

Different behaviour of mouse-human chimeric antibody F(ab')2 fragments of IgG1, IgG2 and IgG4 sub-class in vivo.

Mouse-human chimeric monoclonal antibodies (MAbs) of 3 different human IgG sub-classes directed against carcinoembryonic antigen (CEA) have been produced in SP-0 cells transfected with genomic chimeric DNA. F(ab')2 fragments were obtained by pepsin digestion of the purified chimeric MAbs of human IgG1, IgG2 and IgG4 sub-class and of parental mouse MAb IgG1. The 4 F(ab')2 fragments exhibit similar molecular weight by SDS-PAGE. They were labelled with 125I or 131I and high binding (80 to 87%) to purified unsolubilized CEA was observed. In vivo, double labelling experiments indicate that the longest biological half-life and the highest tumour-localization capacity is obtained with F(ab')2 from chimeric MAb of human IgG2 sub-class, whereas F(ab')2 from chimeric MAb IgG4 give very low values for these 2 parameters. F(ab')2 from chimeric MAb IgG1 and from parental mouse MAb yield intermediate results in vivo. Our findings should help to select the appropriate human IgG sub-class to produce chimeric or reshaped MAb F(ab')2 to be used for tumour detection by immunoscintigraphy and for radioimmunotherapy.