Provided Services

Two ways in which we provide support
There are two ways in which we work with projects, each shown in the figures below:

(Figure 1 coming)

The course presented in figure 1 is typical for projects that have their roots within CTU (i.e., which are largely prepared with support from CTU) and have the same course as, for example, statistical assistance in CTU (Figure 1). This course usually begins with a mapping meeting where we discuss the possibilities for a collaboration and which methods and analyses we suppose might be suitable. We then help to write a protocol that enables us to collect the variables needed for the analysis. In addition, we prepare an analysis plan that indicates how we will carry out the calculations. Before the data collection starts, we look over the database structure together with the data handler to make sure that what is finally exported data matches the structure we need. If we decide to analyse the data that is collected, we sometimes begin to prepare scripts for compilation and transformations of variables during the data collection phase. In some projects, we also create simulation models in this phase that will take the study data as input variables in the analysis. Once the data is collected, the main part of our work begins. As a rule, a shorter period of data cleaning and variable coding is required to structure the data ready for analysis. The analysis plan is then implemented and we create the products the project requires. We often create a technical document that explains how we have carried out the analyses and is a compilation of key information, which is a form of a protocol. This document can thus function as an appendix to scientific publications and is prepared according to a transparent standard that enables an independent analyst to understand and be able to follow the work.

(Figure 2 coming)

Data from projects desiring our assistance without a previous relation to CTU, is typically already collected when we are contacted. Because of this, the course of such projects is somewhat different from previously described. The course for these projects is similar to the roadmap presented in figure 1, starting with a mapping phase where we first discuss whether the data is suitable for a health economic analysis, and further what opportunities the data provides to be able to provide answers to the research questions that are put forward (see Figure 2). If we believe we can carry out the project with good scientific quality, we will prepare an analysis plan. In some projects we are not involved in the design of what data is to be collected and how it may be necessary to collect external data, either from published literature or databases (e.g. health registers, Statistics Norway etc.). In these cases, it will then be necessary to prepare a model that can compare the data from the study with the external data. Projects that have prepared their own databases often require some further processing of the data before the analyses can be carried out.

Projects
We create decision-relevant information in the form of scientific publications with a global audience. We have research expertise and prepare our analyses in accordance with good scientific practice and established guidelines (for example the CHEERS guidelines for publishing cost-effectiveness analyses). In projects with an exclusively Norwegian focus, such as economic analyses of the organisation of health services within Norway, we strive to adhere closely to Norwegian guidelines for method assessments in order to create the most decision-relevant information as possible.