| Methods |
Cluster randomised trial |
| Participants |
Japanese women aged below 35 years with primary subfertility who had been undergoing general fertility treatment for less than 2 years in hospitals and clinics across Japan. Women were excluded if they were candidates for any artificial treatment or if they had an adopted child, foster child, or stepchild.
140 women were randomised into the intervention group (n = 96) or the control group (n = 44). In the intervention group, mean (SD) age was 30.4 (2.87) years and duration of subfertility was 2.3 (1.43) years. In the control group, mean (SD) age was 31.3 (2.49) years and duration of subfertility was 2.2 (1.5) years |
| Interventions |
The intervention group received booklet A during fertility treatment. They read booklet A, did homework assignments related to stress management, used a stress diary, relaxation diary, social support network, and a stress calendar. Once per month, they received feedback by email from the investigator (nurse‐midwife).
The control group received booklet B with a short description of stress management. The control group did not do any homework assignments and did not receive feedback.
The intervention lasted 14 weeks, individual, and by written information |
| Outcomes |
Depression and anxiety (measured by Hospital Anxiety and Depression Scale), general quality of life (measured by SF‐36) and pregnancy rates |
| Notes |
|
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Low risk |
We randomly assigned 7 institutions into an experimental group (4 institutions) and a control group (3 institutions) by using uniform random numbers in Excel |
| Allocation concealment (selection bias) |
Low risk |
Nurses and nurse‐midwives in charge of Recruit and participants were not informed if A and B was experimental group |
| Blinding of participants and personnel (performance bias)
Subjective outcomes |
High risk |
Participants and personnel were not blinded and were likely to be influenced |
| Blinding of participants and personnel (performance bias)
Objective outcomes |
High risk |
No blinding of participants and personnel, objective outcomes could be influenced |
| Blinding of outcome assessment (detection bias)
Subjective outcomes |
High risk |
High risk for blinding of outcome assessor because the participant is the outcome assessor and is not blinded |
| Blinding of outcome assessment (detection bias)
Objective outcomes |
Low risk |
Assessor not blinded, but fewer participants missing from control group. Greater attempts of intervention group follow‐up than of control group follow up, which could lead to bias, will be unlikely in this case |
| Incomplete outcome data (attrition bias)
All outcomes |
High risk |
Withdrawal rate is 31 (> 20% withdrawals) in the intervention group (11 pregnant, 16 did not do their homework, and 4 stopped treatment) versus 6 in the control group (4 pregnant, 1 did not submit questionnaire, 1 was transferred to another clinic). The withdrawal of 16 participants who did not do their homework (the intervention was likely to not have been successful for them) gives an overestimation of the effect in the intervention group |
| Selective reporting (reporting bias) |
Unclear risk |
Unclear without access to the protocol |
| Other bias |
High risk |
There was baseline imbalance in treatment methods between the intervention and control group despite apparently suitable procedures for randomisation and concealment. Validated questionnaires were used for the primary outcomes |
