| Methods |
Design: parallel‐group
Number screened for inclusion: 1,571 (1,230 ineligible: main reasons: lack of menopausal symptoms (293), refusal to stop HT (241), cycle not perimenopausal (206))
Number randomly assigned: 241
Number dropped out: none stated
Number lost to follow‐up: none stated
Number analysed: 211 (30 had data missing from symptom diaries)
Intention‐to‐treat analysis: no
Power calculation:
Duration: two years
Timing: August 1996 to August 1997
Location: Wake University Clinic,Carolina, USA
Funding: soy supplements supplied by industry (Soy Technologies, St Louis, Missouri, USA) |
| Participants |
Inclusion criteria: perimenopausal women (no more than one menstrual period in three months before randomisation), at least one vasomotor symptom per day, not using HT for three months before recruitment, willingness to participate in one‐week run‐in with isoflavone‐free supplement
Exclusion criteria: acute MI or stroke within previous six months, history of breast or endometrial cancer, invasive cancer within previous five years, active thromboembolic disease, previous osteoporosis‐related fractures treated with hormones, low baseline bone density, previous exposure to diethylboestrol, dyslipidaemia, endometrial biopsy showing hyperplasia, consumption of soy products on a daily basis and unwillingness to reduce consumption to once a week
Age, years: mean 50.8 (SE 0.2)
Recruitment method: "recruited from the community" |
| Interventions |
Dose, duration and timing of administration: one 25 g ready‐to‐drink beverage daily, chocolate or orange flavoured, for two years
|
| Outcomes |
Menopausal symptoms: hot flushes, night sweats recorded in monthly calendar with daily entry field: participants asked to record number and severity of symptoms for one full week per month
Compliance: compliance calendars |
| Notes |
37 women (18%) took HT during trial: 16 in high‐dose group (25%), 11 in middle group (16%), 10 in control group (13%). Data analysed with and without these women, and pattern of results not affected |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Unclear risk |
Not stated |
| Allocation concealment (selection bias) |
Unclear risk |
Not stated |
| Blinding of participants and personnel (performance bias)
All outcomes |
Low risk |
Stated as double‐blind |
| Blinding of outcome assessment (detection bias)
All outcomes |
Low risk |
Stated as double‐blind |
| Incomplete outcome data (attrition bias)
All outcomes |
Unclear risk |
Moderate dropout: missing data from questionnaires because not filled in |
| Selective reporting (reporting bias) |
High risk |
Authors did not report on all prespecified outcomes |
