2002 International International Biennial Scientific Meeting

PREDICTING NON-PUERPERAL EPISODES OF ILLNESS IN WOMEN WITH BIPOLAR AFFECTIVE PUERPERAL PSYCHOSIS

Emma Robertson, Ian Jones and Nick Craddock

Molecular Psychiatry Group, University of Birmingham, Department of Psychiatry, Queen Elizabeth Psychiatric Hospital, Edgbaston, Birmingham

e.k.robertson@bham.ac.uk

Objectives:

• To ascertain the rate of non-puerperal episodes of illness in 110 women diagnosed with puerperal psychosis.
• To conduct survival analysis to examine whether clinical features influence the timing or outcome of non-puerperal episodes.

Introduction: Puerperal psychosis is the most severe and rare of postnatal affective illnesses, affecting 1-2 per 1000 deliveries. There is good evidence from clinical, outcome and genetic studies that most cases of puerperal psychosis are bipolar in nature. Research studies have shown that women who experience puerperal psychosis are at high risk of experiencing further puerperal and non-puerperal episodes of illness. Estimates of non-puerperal relapse of affective illness vary from 39% - 81%, however it has been proposed that these relapse rates may be influenced by certain clinical factors. This study aimed to establish relapse rates of non-puerperal illness in a large sample of women diagnosed with puerperal psychosis.

Methods: 110 women who had experienced at least 1 episode of mania or psychosis, according to DSM-IV criteria, within six weeks of childbirth were included. Their illness histories were charted to establish the number and timing of non-puerperal episodes of illness.

Results: 63% of women suffered from non-puerperal relapse of illness following an episode of puerperal psychosis.

Using survival analysis, two hypotheses were tested. Firstly, that women with a previous history of psychiatric illness would have a shorter time to non-puerperal relapse. The findings showed that the median time to relapse for women with a previous history was 3 years compared to 6 years for women with no previous history (p = 0.058).

Secondly, women with a positive family history of mental illness would have a shorter time to non-puerperal relapse than women without a family history. The results showed that the median time to relapse for women with a family history was 3 years compared to 8 years (p <0. 0001).

Conclusions: This study showed two clinical factors appear to influence the timing of non-puerperal episodes of illness. Whilst knowing the rates of non-puerperal illness is useful, further work examining factors that may influence the timing of episodes has great implications in counselling women with regard to the risk of further non-puerperal illness episodes.