Older age mothers (1995-9)

The lifetime experiences of an older age group of mothers was examined to look both at experiences of a previous cohort of mothers, for example those born before the second world war, but also to examine the experience of mothers whose daughters had proved high risk for depression in midlife. Establishing how risk may have been transmitted intergenerationally for this age group would add to the lifespan models developed. Such investigation is of course dependent upon good retrospective measures, and since the CECa and ALPHI interview had proved so robust on midlife women, it was hoped that the same would hold for an even older age group. In addition, social isolation has been identified as a risk in older persons for depression and anxiety orders. ‘Disengagement’ has been viewed as a characteristic of older life stages and explained in terms of loss of close others with low opportunity for increasing attachment. Therefore attachment style in older persons, specifically around avoidance, was examined as a theoretical framework for exploring isolation in relation to disorder.



To look at lifetime psychosocial risk for disorder in an older group of mothers aged 50-85. To test the same model as in midlife women, investigating childhood experience, adult adversity, attachment style and self-esteem. Also, to compare risk in these mothers and their midlife daughters to look for mechanisms of intergenerational transmission.


The sample

81 women aged between 50 and 75 were selected for interview. These comprised three groups:

Intergenerational high-risk series - 33 mothers of midlife women already studied who had been selected on the basis of GP screening in 1990-5 as vulnerable to depression. The daughters all had psychosocial risks in terms of either childhood neglect or abuse, or poor adult relationships. This included a series of sisters.
Intergenerational comparison series - 23 mothers of midlife comparison women who were selected consecutively from GP surgery records and studied in 1990-5. Each mother had two daughters in the study.
New series - 25 newly selected women from GP lists, all of whom had a midlife daughter in Islington in common with the other two groups.

9% of the older mothers directly refused the interview and 11% of midlife daughters refused to contact their mothers for the study. A further 20% were either too ill (7%) or had died (13%) since the original study. 13% of either midlife daughters or their mothers proved unobtainable at the follow-up of the original study (eg having moved away). 13% of mothers were not contacted because they did not bring up the daughter. This resulted in 34% of the original mothers identified being interviewed.

Of the newly selected series: 108 women were screened by questionnaire through GP surgery listings. 9% refused to be interviewed, 7% proved unobtainable, 22% were unsuitable having no midlife daughter in Islington and 23% were interviewed.

For other analyses midlife daughter and older mothers were paired to look at intergenerational transmission. For this analysis all 81 women were analysed as a single group to examine inter-respondent risks for disorder and social support.

There were no significant differences between the three groups in demographic or disorder variables.



The 81 community-based mothers aged between 50 and 75 (average age 64) were selected for interview. All women were contacted for interview in their own homes, wherever they lived in the UK. The ‘mini-mental’ test was used to screen out women over 65 with cognitive deficit or signs of dementia. A lifespan interview covering childhood and adult adversity, current support and attachment style and lifetime episodes of depression and anxiety was conducted.


Measures used

  • Childhood Experience of Care and Abuse (CECA) interview
  • The Adult Life Phase Interview (ALPHI)
  • Parenting
  • Attachment Style Interview (ASI) and Self-esteem
  • SCID for lifetime disorder


Disorder: 24% of women experienced case of depression or anxiety in the 12 months before interview and 17% had chronic or recurrent episodes during their lifetime. This was a lower rate than shown by their daughters who had been studied, and substantially lower than the adolescents studied.

Attachment style: Three-quarters of the women were judged to have highly insecure attachment style, of which ‘Withdrawn’ was the most common. Insecure attachment style was significantly related to social isolation, marked loneliness and history of poor partner relationships. However, consistent with other analyses, Withdrawn style was unrelated to affective disorder, Fearful or Angry-Dismissive styles had the highest rates of disorder. Childhood neglect/abuse and adult adversity were associated with increased rates of such insecure attachment. Attachment style is an important aspect of older life experience and can denote isolation, loneliness and poor support. This needs to be understood in treating older persons at risk for affective disorder. Attachment style proved to be associated with disorder as in all series including the elderly. Rates of insecure were high in the elderly due mainly to paucity of close relationships with the most common attitudes denoting 'withdrawn avoidance' indicating high self-reliance and attitudinal constraints on closeness and help-seeking.

Lifetime psychosocial risk in the form of childhood neglect/abuse, adult attachment style and chronic adversity are associated with affective disorder in this series as in all the three generations. These replicate previous findings on the mid-life series studied in the last programme. Associations with childhood experience were however, weakest in the elderly series. This maybe due to the weakening influence of childhood with the passage of time, or may be due to a different environmental impact and sensitivity on an earlier cohort. These alternative explanations require further analysis. Young daughters were found to have significantly more adult life phases and change points than women in their grandmothers' generation, implying an acceleration of pace of life in terms of change of partners, living place and major roles. Where such change is associated with premature embarkation into adult roles this may account for some of the increase in risk. Further analysis of adversity and change will be made to test such hypotheses.

Inter-generational analysis: Characteristics of older age mothers and their midlife daughters was examined in terms of two outcomes in the offspring: neglect/abuse in childhood and lifetime recurrent disorder. Mothers insecure attachment at interview was highly related to disorder in her daughter at follow-up. There was however no relationship between mothers depression (either in 12 months before last contact or lifetime recurrent disorder), or fathers lifetime disorder and the daughters disorder. Mother's own neglect or abuse in childhood was unrelated to her daughter's neglect/abuse in this high risk series although it was significant at the 5% level in the comparison series. Mothers, but not fathers, lifetime chronic or recurrent disorder was significantly associated with her daughters experience of neglect/abuse Logistic regression showed that daughters neglect/abuse and mothers vulnerability were required to model daughters disorder. Lifetime disorder of either mother or father did not add to the model.Logistic regression showed that daughter's neglect/abuse in addition to mothers non-standard attachment style was required to model daughters disorder.



Five hundred people recall their life story… all kept  in one collection


Memories of childhood and of adult life: adversity, support relationships ...

Reports of coping style, self esteem,
relating styles,
psychological disorders ...