ABSTRACT
The dataset to be analyzed in this chapter originates from a clinical trial of the use of estrogen patches in the treatment of postnatal depression; full details are given in Gregoire et al. (1996). In total, 61 women with major depression, which began within 3 months of childbirth and persisted for up to 18 months postnatally, were allocated randomly to the active treatment or a placebo (a dummy patch); 34 received the former and the remaining 27 received the latter. The women were assessed pretreatment and monthly for six months after treatment on the Edinburgh postnatal depression scale (EPDS), higher values of which indicate increasingly severe depression. The data are shown in Table 8.1; a value of −9 in this table indicates that the observation is missing. The non-integer depression scores result from missing questionnaire items (in this case the average of all available items was multiplied by the total number of items). The variables are
subj: patient identifier group: treatment group (1=estrogen patch, 0=placebo patch) pre: pretreatment or baseline EPDS depression score
dep1 to dep6: EPDS depression scores for visits 1 to 6 The main question of interest here is whether the estrogen patch is effective at reducing post-natal depression compared with the placebo.
The data in Table 8.1 consist of repeated observations over time on each of the 61 patients; such data are generally referred to as longitudinal data, panel data or repeated measurements, and as cross-sectional time-series in Stata. There is a large body of methods that can be used to analyze longitudinal data, ranging from the simple to the complex. Some useful references are Diggle et al. (2002), Everitt (1995), and Hand and Crowder (1996). In this chapter we concentrate on the following approaches:
