Venn 1999.

Methods	Retrospective cohort.
Participants	The cohort consisted of 29,700 women who registered with at least one of 10 participating Australian IVF clinics before Jan 1, 1994. Women whose usual residence was outside Australia (n = 623), or whose date of birth or age was unknown (n = 520) were excluded. Exposed were 20,656 women who had at least one IVF treatment cycle with ovarian stimulation to induce multiple folliculogenesis, including stimulated cycles that were cancelled before oocyte collection. Unexposed were considered 9044 women who registered for IVF but did not receive treatment or had 'natural' cycle treatment without ovarian stimulation. Four women with cancer diagnosis before start of IVF treatment were excluded. Follow‐up was from the time each woman entered the cohort until the first of: date of cancer diagnosis, death, or Dec 31 of the year of complete cancer data for her state of residence. Median follow‐up of the exposed cohort, 7 years; of the unexposed cohort, 10 years.
Interventions	At least one IVF treatment cycle with ovarian stimulation to induce multiple folliculogenesis. Drugs used were clomiphene citrate, human menopausal gonadotropin, both clomiphene citrate and human menopausal gonadotropin, human menopausal gonadotropin and gonadotropin‐releasing hormone agonist. Data on number of fertility treatment cycles were available.
Outcomes	Endometrial cancer ascertained by record linkage to population‐based cancer registries held by the state or by the National Cancer Statistics Clearing House and to the National Death Index. Incident cases in cohort, 12; in exposed cohort, 5.
Notes	Subfertility investigations routinely used by IVF clinics and referring gynaecologists included hormone assays, ultrasonography, diagnostic laparoscopy, and semen analysis. The cause of subfertility was classified as tubal, male factor, endometriosis, ovarian disorders (including ovulation disorders, polycystic ovary syndrome, premature menopause, and oophorectomy), other factors (including cervical factors, other uterine abnormalities, donor egg recipients for genetic disease, altruistic egg donors), or unexplained.
Risk of bias
Bias	Authors' judgement	Support for judgement
Selection bias (comparability)	High risk	Quote: "Any diagnosis of cancer immediately before a woman’s registration with an IVF programme was checked to confirm that cancer preceded referral for IVF. Four women with cancer were referred for IVF with the aim of producing embryos for freezing before cancer therapy started; they were excluded from the analysis". Comment: Outcome not likely to be present at start. Quote: "Women classified as unexposed were those who registered for IVF but did not receive treatment (95%) or who had 'natural' cycle treatment without ovarian stimulation (5%)". Comment: Non‐exposed drawn from the same population as the exposed cohort. Comment: No comparability on cause of subfertility, diabetes mellitus, polycystic ovary syndrome (PCOS), and obesity was ensured.
Selection bias (confounding)	High risk	Quote: "For the SIR, the age‐specific rates for the three cancers were compared across three calendar periods (1981–85, 1986–90, and 1991–95) and across the states of Australia". Comment: Analyses adjusted only for age.
Performance bias	High risk	Quote: "Data collected from clinics were woman’s name and date of birth, partner’s name, address, date of registration with the clinic, and cause of subfertility. Information about IVF treatment consisted of dates of treatment cycles, type of treatment (e.g. stimulated cycle or frozen embryo transfer), fertility drugs used for ovarian stimulation, number of oocytes collected, and whether the cycle was cancelled before oocyte collection". Comment: Exposure to ovary‐stimulation drugs was ascertained by a secure source, namely, medical records. Blindness regarding the allocated interventions not guaranteed.
Detection bias	Low risk	Quote: "Ascertainment of cancer cases was by record linkage to population‐based cancer registries held by the state or by the National Cancer Statistics Clearing House and to the National Death Index". Comment: Ascertainment of outcome by record linkage.
Attrition bias	Low risk	Quote: "Although the registries covered all states and territories in Australia, and loss to follow‐up outside the country is likely to have been negligible, we may have missed women with cancer who had changed their names since participating in an IVF programme. We have estimated that more than 25% of breast cancers would have had to be missed for the SIR to be greater than 1·0 in this IVF cohort—an unlikely occurrence". Comment: Follow‐up was expected to be rather complete.
Selective reporting (reporting bias)	Low risk	Comment: "All of the study’s prespecified (primary and secondary) outcomes that were of interest in the review have been reported in the prespecified way".
Other bias	High risk	Quote from Table 1: "Median duration of follow‐up in exposed women: 7 years". Comment: Inadequate length of follow‐up (< 10 years). Comment: Non‐RCT study.