Cognitive processing of infant stimuli in pregnant women with and without affective disorders and the association to postpartum depression

Multivariable regression models are powerful tools that are used frequently in studies of clinical outcomes. These models can use a mixture of categorical and continuous variables and can handle partially observed (censored) responses. However, uncritical application of modelling techniques can result in models that poorly fit the dataset at hand, or, even more likely, inaccurately predict outcomes on new subjects. One must know how to measure qualities of a model's fit in order to avoid poorly fitted or overfitted models. Measurement of predictive accuracy can be difficult for survival time data in the presence of censoring. We discuss an easily interpretable index of predictive discrimination as well as methods for assessing calibration of predicted survival probabilities. Both types of predictive accuracy should be unbiasedly validated using bootstrapping or cross-validation, before using predictions in a new data series. We discuss some of the hazards of poorly fitted and overfitted regression models and present one modelling strategy that avoids many of the problems discussed. The methods described are applicable to all regression models, but are particularly needed for binary, ordinal, and time-to-event outcomes. Methods are illustrated with a survival analysis in prostate cancer using Cox regression.

An eleven item clinician-administered Mania Rating Scale (MRS) is introduced, and its reliability, validity and sensitivity are examined. There was a high correlation between the scores of two independent clinicians on both the total score (0.93) and the individual item scores (0.66 to 0.92). The MRS score correlated highly with an independent global rating, and with scores of two other mania rating scales administered concurrently. The score also correlated with the number of days of subsequent stay in hospital. It was able to differentiate statistically patients before and after two weeks of treatment and to distinguish levels of severity based on the global rating.