Prospective study of symptoms of ovarian cancer in postmenopausal women

• Main Author: Allen, P. L.
• Published: University College London (University of London) 2010
• Subjects: 610
• Online Access: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.564977

Background: Research has identified key symptoms of ovarian cancer, although there are gaps in the knowledge about the pattern of symptom onset, severity and frequency. Previous studies are limited by use of non-validated questionnaires, recall bias and under-reporting bias in medical records. Aim: The aim of the research was to prospectively identify type, severity, frequency and duration of symptoms that precede ovarian cancer diagnosis in postmenopausal women. Methods: Questionnaire development methods described by the European Organisation for Research and Treatment of Cancer were utilised to develop a validated ovarian cancer symptoms questionnaire (OCSq). Interviews with 21 gynaecological oncology clinicians/nurses and 25 women with ovarian cancer guided development of the OCSq. The OCSq was piloted among 1,339 women and posted to 100,000. Results: A total of 829 women completed a pilot OCSq and baseline analysis of the finalised OCSq included 51,007. Symptoms were ubiquitous, with 89% of women reporting any symptoms, 55-56% symptoms at level 2-3 severity and 42-49% at ≥12 days frequency and <12 months duration. Abdominal/pelvic pain, increased abdominal size/bloating or feeling full at ≥12 days and <12 months was reported by 11-16%. There were 263 women who had an abnormal ovarian cancer screening result in the pilot and two women were diagnosed with ovarian cancer. No symptoms were consistently associated with abnormal results when severity, frequency and duration criteria were added to analyses. Multivariate analyses found age, pelvic pressure, tiredness/fatigue, pelvic bloating/fullness, shortness of breath, leg ache/pain and abdominal pressure independently predicted abnormal ovarian cancer screening results at various levels of analysis. However, odds ratios were low and confidence limits were wide. Symptom reporting was strongly correlated with previous awareness of the possibility of an ovarian lesion and depression screening status. Conclusion: The research is currently ongoing with follow-up analyses planned to commence in late 2010. Preliminary findings indicate that there is currently insufficient evidence to justify symptoms awareness campaigns based upon the results of previous retrospective research, and that such campaigns risk overwhelming primary care services with ‘worried-well’ women, increasing psychological morbidity, service costs, unnecessary investigations and potentially harmful surgery.