Study characteristics
Of 791 studies identified in our search, 15 met inclusion and exclusion criteria and were included for analysis (Figure 1 ).17-32 Five studies were excluded as they either did not provide data in an extractable format or used language spoken at home or country of birth as a surrogate for ethnicity.33-37 Characteristics of the included studies are presented in Table S1 . These fifteen studies encompassed 72,555 stillbirths, equating to a stillbirth rate ranging from 0.35% to 6.8% of all births. Seven were conducted in the United States of America, five in Australia, and three in the United Kingdom. Most (n = 10) were conducted more than 10 years ago and of these, six had data collected more than 20 years ago. Three studies were limited to a single institution, seven were multicentred, and five were conducted at a population or national level. Only eight of the included studies provided an indication of the autopsy rate. In these studies, the autopsy rate ranged from 37.6% to 100%. While the definition of stillbirth varied slightly between studies, a majority of studies defined stillbirth as death of foetus of at least 20 weeks’ gestation. The classification system used to describe the cause of stillbirth varied widely, with four studies using the ICD system, three studies using INCODE, two studies using PSANZ, and six using a classification system unique to the study. Most studies were either retrospective cohort studies (n = 7) or cross-sectional in nature (n = 4), three were case-control, and one was prospective. Only one study was deemed to be of high quality; the rest were deemed to be either medium (n = 7) or low (n = 7) quality (Table S2 ).