Kashimura 1986

Methods	RCT conducted in Japan. Participants recruited between 1963 and 1977. Stated as a prospective study with participants 'selected at random' to receive prophylaxis This may not be a true RCT
Participants	420 women with molar pregnancy (low and high risk) Excluded women who were referred longer than 3 weeks after evacuation, those who had received other drugs for prophylaxis (see 'Risk of bias' table below), women who had undergone hysterectomy and women diagnosed as having partial mole or hydropic degeneration
Interventions	Arm 1: methotrexate 10 mg daily (IM or oral) for 7 days, within 3 weeks of evacuation (293 women) Arm 2: no P‐Chem (127 women) Women were followed up weekly with urine hCG measurements
Outcomes	GTN diagnosed by histology or Ishizuka score (a risk rating system used in Japan); side effects and subsequent pregnancy
Notes	5‐ to 15‐year follow‐up reported. Time to invasive mole diagnosis was 56.8 days in P‐Chem group and 42.7 days in control group (SD not given; P = 0.6). No attrition occurred for primary outcomes. Only reported adverse effects in the P‐Chem group: 27.3% experienced drug‐related side effects including stomatitis (10.3%), nausea/vomiting (6.8%) and leukopenia (4.4%). Grades of toxicity were not reported but the report states that there were no severe complications or drug‐related deaths. Baseline characteristics were not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	'Selected at random' suggests that this was not a truly random process. No randomisation ratio was described and yet the 2 groups were very different sizes (293 vs. 127). The authors stated that 39 patients who received other drugs besides methotrexate were excluded from the study; this also suggests a non‐random study design
Allocation concealment (selection bias)	Unclear risk	Not described
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not described
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not described
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Attrition was low for the main outcomes but high for long‐term outcomes such as subsequent pregnancies
Selective reporting (reporting bias)	High risk	Baseline characteristics of the 2 groups were not compared/reported
Other bias	High risk	More women in the control group were ≥ 40 years old (22% vs. 11%) and progressed to GTN (39% vs. 12%)