Thesis (Ph. D.)--University of Oregon, 2006. / Typescript. Includes vita and abstract. Includes bibliographical references (leaves 127-130). Also available for download via the World Wide Web; free to University of Oregon users.

Health and public health services in England are under increasing financial pressure. At the same time, nearly 40% of the total disease burden is potentially amenable to known causes with two of the leading behavioural risk factors being unhealthy diets and physical inactivity. To better inform decision makers and improve health in England, this thesis aims to develop a cost-effectiveness model that can directly compare diet and physical activity interventions. Published public health economic models were reviewed and the strengths and weaknesses of the modelling structures were explored. A pre-existing multistate life table model, PRIMEtime, was developed into PRIMEtime Cost Effectiveness (PRIMEtime CE). Disease specific NHS England costs were derived from NHS England Programme Budgeting Data and unrelated disease costs from NHS cost curves. Social care costs were quantified using a Department of Health tool for estimating wider societal costs. Disease specific utility decrements were adopted from a catalogue of EuroQoL five dimensions questionnaire scores. The cost effectiveness of reformulating food to have less salt and of expanding access to leisure centres in England were modelled from an NHS and social care perspective over a 10 year time horizon, including government and industry costs. Salt reformulation was dominant with an estimated cost per quality adjusted life year (QALY) of -£17,000 (95% uncertainty interval, -£40,000 to £39,000), compared with £727,000 (£514,000 to £1,064,000) for increasing access to leisure centres. Sensitivity analyses and cross validation testing of outcomes demonstrated how cost per QALY estimates were sensitive to the choice of model scope, parameters, and structure. PRIMEtime CE is a tool for decision makers to compare interventions affecting diet and physical activity, enabling them to make better informed choices about how to spend finite resources. Future work will focus on making the model freely available and expanding its risk factors to enable comparisons of other public health interventions.

Die Palliativversorgung (PV) verfolgt das Ziel, die Lebensqualität in der noch verbleibenden Lebenszeit von Patienten mit lebensbedrohlichen Erkrankungen zu maximieren. Deutschland verfügt mittlerweile über ein umfassendes PV-System im ambulanten und stationären Sektor und einen Anspruch auf PV als Teil der Regelversorgung. Im Rahmen dieser Untersuchung soll überprüft werden, ob die Inanspruchnahme der PV-Institutionen der vom Gesetzgeber und medizinischen Experten intendierten Reihenfolge entspricht und ob es Überlebenszeit- oder Gesundheitsausgabenunterschiede gibt (jeweils im Vergleich zu Nicht-Palliativpatienten). Dazu wird ein Paneldatensatz der AOK PLUS (Sachsen und Thüringen) mit Patienten verwendet, die zwischen 2009 und 2012 an einer Krebserkrankung litten (n=447.191). PV-Patienten werden entsprechend ihres Inanspruchnahmeverhaltens in vier Interventionsgruppen eingeteilt, von denen jeder mittels Propensity Score Matchings eine eigene Kontrollgruppe zugeordnet wird. Als statistische Werkzeuge werden v.a. Übergangswahrscheinlichkeiten, Kaplan-Meier-Überlebensfunktionen sowie lineare und nicht-lineare Regressionsmodelle verwendet. Die Ergebnisse legen nahe, dass die Reihenfolge der Inanspruchnahme im Einklang mit Gesetzen und Richtlinien ist. Überlebensnachteile der PV-Patienten können nicht festgestellt werden. Die Gesundheitsausgaben steigen nach erstmaliger Inanspruchnahme einer PV-Institution in allen Stichproben stark an. Dieses Ergebnis ist sehr robust gegenüber Änderungen der Modellspezifikation, des Modelltyps und der Stichprobe. Die Ergebnisse lassen auf eine hohe Struktur- und Prozessqualität der PV-Angebote schließen. Jedoch führt die Inanspruchnahme von PV in ihrer derzeitigen Form offenbar nicht zu Einsparungen. Ein weiterer Ausbau des PV-Systems finanziert sich demnach nicht „von selbst“. Aufgrund der sehr kurzen Nachbetrachtungszeiträume bleibt die gesundheitsökonomische Analyse der PV weiterhin herausfordernd. / Palliative Care (PC) is an approach for patients with life-threatening diseases that focuses on improving quality of life rather than maximizing the remaining life time. Meanwhile, Germany possesses a comprehensive PC system in the ambulatory and inpatient sector where PC treatments are part of standard care. The objective of this research is to evaluate whether patients are using PC institutions as intended by law and medical experts and whether PC patients differ in terms of survival time or health care expenditures (HCE) in comparison to non-PC patients. For this purpose, a panel data set from the statutory health insurance AOK PLUS (covering the German federal states Thuringia and Saxony) is used, that includes all deceased cancer patients between 2009 and 2012 (n=447,191). According to their usage of PC institutions, PC patients were grouped into four different intervention groups and thus each of them was paired with a control group derived from a propensity score matching. A variety of statistical tools has been used, e.g. transition probabilities, Kaplan-Meier survival functions as well as linear and non-linear regression models. Results show that the intended sequences of PC usage are in accordance with law and medical guidelines. There are no disadvantages in survival of PC patients. In all four samples, HCE of PC patients are higher after the initial contact with a PC institution. This result is very robust against adjustments to the model specification, the model type and the sample. Results suggest that structural and process quality of PC is high. However, no saving effect can be identified for PC in its current form. A further extension of the PC system is therefore not “self-financing”. Due to the very short post treatment time, health economic analysis of PC remains challenging.