Long-term health status and trajectories of seriously injured patients: A population-based longitudinal study

Relative risk is usually the parameter of interest in epidemiologic and medical studies. In this paper, the author proposes a modified Poisson regression approach (i.e., Poisson regression with a robust error variance) to estimate this effect measure directly. A simple 2-by-2 table is used to justify the validity of this approach. Results from a limited simulation study indicate that this approach is very reliable even with total sample sizes as small as 100. The method is illustrated with two data sets.

It has become increasingly common for preference-based measures of health-related quality of life to be used in the evaluation of different health-care interventions. For one such measure, The EuroQol, designed to be used for these purposes, it was necessary to derive a single index value for each of the 243 health states it generates. The problem was that it was virtually impossible to generate direct valuations for all of these states, and thus it was necessary to find a procedure that allows the valuations of all EuroQol states to be interpolated from direct valuations on a subset of these. In a recent study, direct valuations were elicited for 42 EuroQol health states (using the time trade-off method) from a representative sample of the UK population. This article reports on the methodology that was adopted to build up a "tariff" of EuroQol values from this data. A parsimonious model that fits the data well was defined as one in which valuations were explained in terms of the level of severity associated with each dimension, an intercept associated with any move away from full health, and a term that picked up whether any dimension in the state was at its most severe level. The model presented in this article appears to predict the values of the states for which there are direct observations and, thus, can be used to interpolate values for the states for which no direct observations exist.