when the child was 4 years old (Cogill et al., 1986). Stein et al. (1991) found that 19-month-olds whose mothers had recovered from depression that had occurred during the first postnatal year showed lower quality interaction with their mothers and with a stranger than did children whose mothers had never been depressed.

There are some exceptions to this finding, and these are intriguing. For example, Field (1992) reported that 75 percent of mothers who had been depressed early in the postpartum period continued to have symptoms at 6 months postpartum. The infants of the remaining 25 percent did not display a depressed style of interaction or have lower Bayley mental and motor scale scores at 1 year of age (Field, 2002).

Another intriguing question that has been addressed by some of the longitudinal studies is: If parenting quality improves with remission of depression, do children benefit? A few studies help to answer that question. Campbell, Cohen, and Meyers (1995), for example, found that mothers who were depressed 2 months postpartum but whose depression remitted by 6 months were significantly more positive and more competent in feeding their infants relative to mothers whose depression was chronic through 6 months postpartum (Campbell, Cohn, and Meyers, 1995). Furthermore, the infants in the depression remission group were significantly more positive in face-to-face interactions with their mothers than were those whose mothers remained depressed, although they did not differ significantly in terms of negative interaction or in the quality of engagement with their mothers in toy play.

Some studies that included an active treatment component for the mothers’ depression also included mother-infant assessments, providing an opportunity to more directly test the hypothesis that parenting is one of the mechanisms in the transmission of risk from depressed mothers to their children. Treatment studies are reviewed in Chapter 6, but we focus here on the small subset of treatment studies that allow us to address this issue. A controlled trial of interpersonal psychotherapy in postpartum women with major depression, which was found to be effective in reducing levels of depression, also found significant improvement on self-reported measures of mothers’ relationships with their children associated with interpersonal psychotherapy, even though the women did not achieve the levels typical of women with no history of depression (O’Hara et al., 2000). More recent studies continue to show that despite improvement in depression with interpersonal psychotherapy, mother-infant relationships were not improved (Forman et al., 2007).

Similarly, Cooper and Murray (1997), with a community sample screened for depression, found that treated mothers (randomly assigned to either nondirective counseling, cognitive-behavioral therapy, or dynamic psychotherapy), despite significant improvement in mood, were not ob-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement