Scandinavian journal of gastroenterology. Supplement
-
Scand. J. Gastroenterol. Suppl. · Jan 1988
Comparative StudySingle case studies from a statistician's point of view.
In a clinical single case study the object is to decide whether a certain treatment has an intended effect on a certain patient. It is pointed out that the design should only be used in clinical situations where several rather strong assumptions are fulfilled. However, in these situations the design may be of significant value.
-
The limitations of conventional group comparative therapeutic trials are discussed. They include: the heterogeneity problem (due to biological variations within the sample) and the extrapolation problem (i.e. the problem of external validity). These problems may to some extent be overcome by multiple cross-over studies in a single patients. ⋯ Usually, it is necessary to measure the clinical effect by means of clinical scores, for which reason ranking methods must be used. The statistical significance may be judged by means of a permutation test. The risk of committing a type II error in single patient studies is usually high.
-
Scand. J. Gastroenterol. Suppl. · Jan 1988
Comparative StudyExperience with a multi crossover model in dyspepsia.
A multi cross over model (MCOM) has been designed for single case studies. The model which is only partly randomized, implies regular interchanges between treatment periods with active drug and placebo. ⋯ The advantages of the single case approach and the impact of the MCOM is illustrated by the results from a study of the effect of cimetidine in non ulcer dyspepsia. Although there are several statistical objections to the model, the results from the study are reasonable and demonstrate a small degree of violence of preassumptions.
-
Scand. J. Gastroenterol. Suppl. · Jan 1988
Comparative StudyStatistical issues in studies of individual response.
We consider intensive studies of individual response to therapy in a controlled experiment. It is helpful to distinguish between strict 'N = 1' studies, which are pragmatic trials intended to draw conclusions concerning only the patient under consideration, and 'N much greater than 1' studies, which are explanatory trials intended to make more general statements about a treatment with highly variable response, for which aggregate measures of effect on groups are inappropriate. Issues of design, measurement, and statistical significance are discussed, and the rule of permutation tests emphasised. Three published examples are used as illustrations.