It was concluded that there is now sufficient evidence that clinicians should not assume the EPDS to be unidimensional, but should assess all three subscales when screening for susceptibility to post-partum depression and/or post-partum anxiety.

This research examined experiences of pregnancy and delivery as predictors of three subscales identified within the Edinburgh Postnatal Depression Scale (nonspecific depression, anhedonia, and anxiety). Mothers of babies under 1 year (N = 415) completed the Edinburgh Postnatal Depression Scale (EPDS) and gave information as to number of pregnancies, number of children, and their ratings of the difficulty of the course of pregnancy and delivery. Number of pregnancies and number of children did not predict scores on any of these subscales. Difficulty in the course of pregnancy was a significant positive predictor of nonspecific depression, and difficulty of delivery was a significant positive predictor of anxiety. Neither of these variables significantly predicted anhedonia. These results provide evidence of the heterogeneity of postpartum depression, indicating that risk factors differentially influence subscales of the EPDS.

Factor analysis of the Edinburgh Postnatal Depression Scale (EPDS) yielded three subscales (i.e. symptoms of, respectively, nonspecific depression, anhedonia, and anxiety). Each of these was regressed on three known risk factors for postpartum depression (change in quality of marital relationship, social support availability, and social support satisfaction). All three risk factors were significant independent predictors of nonspecific depression. Change in quality of marital relationship, but not social support availability or satisfaction, was a significant predictor of anhedonia; and none of the three predicted anxiety. These results support the view that the EPDS is not a unidimensional instrument, and that its separate components are differentially affected by various risk factors.

In a cross-sectional study five groups of male recruits (n = 369; ages 21 to 36) to a large English provincial police force were tested: one group at entry, three groups at different stages of their 2-year probationary period (one at one-quarter, one at one-half and one at three-quarters of the way through) and one immediately after the probationary period. Significant group effects were found on all 27 items of a set of self-concept statements, suggesting effects of entry-level occupational socialization on recruits' self-image. Changes from group to group during the course of training were examined using discriminant function analysis. Factor analysis yielded four clusters of items: isolation/vulnerability, tension/apprehension, self-worth and commitment/empathy; scores on all four fluctuated during training. Isolation and apprehension rose from a low base during training, and self-worth fell from a high base, but entry levels were recovered by completion of training, giving support for a weak version of the U-curve hypothesis of adaptation. However, high levels of commitment and empathy at entry were significantly depressed on completion of training. In addition, older trainees reported lower self-worth and higher apprehension. Results are interpreted in terms of adaptation during occupational socialization towards a more instrumental and cynical operational style.

This study addressed the effects of age on anxiety and depressive symptoms. The analysis was based on the responses of 1,334 retired male Scottish police officers (34-94 years old) to the Hospital Anxiety and Depression Scale. Multiple regression analysis was used to determine the partialed linear and curvilinear effects of symptoms of, separately, anxiety and depression on age and retirement variables. Significant partialed effects of retirement type and present age were found. Early retirement was a vulnerability factor for both disorders. The authors found a consistent linear reduction in anxiety across age and a U-shaped function for depression across age. The differing profiles and independent effects of age on anxiety and depression are consistent with their status as separate entities. Respondents were increasingly susceptible to depression from the mid-50s onward, whereas susceptibility to anxiety was reduced with age.