Despite the global decline in smoking prevalence, tobacco smoking remains high in many countries. 1 One such country is China where WHO estimates that, in 2010, 51% of men and 2% of women aged 15 years and over smoked. 2 It is also projected that by 2025, the figure for men will have only decreased to 46% and will remain at 2% for women. 2 Given the known association between maternal smoking in pregnancy and an array of adverse outcomes including miscarriage, fetal growth restriction, preterm births and perinatal mortality, the very low rates of smoking among women in China are welcome. However, another important consideration is the impact of exposure to second-hand smoke (SHS), exhaled smoke and the smoke from the lit end of the cigarette, which is implicated in 1% or over 600 000 deaths per year worldwide. 3 So even in countries with low smoking rates among women, exposure to SHS remains high and of concern. Exposure to SHS will have the largest impact in low-to middle-income countries due to overcrowding, poor ventilation and high levels of household smoking.Despite the evidence suggesting that exposure to SHS in pregnancy can result in adverse pregnancy outcomes, 4 associations with some pregnancy outcomes including congenital anomalies, which affect 2%-4% of all pregnancies, remain less clear. A systematic review of observational studies reported a significant increased risk of congenital anomalies with SHS exposure (OR 1.3 [95%CI 1.01, 1.276]) although associations with individual congenital anomalies remained weak. 5 Since exposure to SHS is potentially a modifiable risk factor associated with adverse pregnancy outcome, further evidence on its impact is warranted.In this issue of Paediatric and Perinatal Epidemiology, Pi et al 6 report data on the association of exposure to SHS during the periconceptual period (1 month before to 2 months after conception) among non-smoking women with an orofacial cleft (OFC) affected pregnancy (240 women and 1420 women with a non-OFC affected pregnancy). The study was based on data collected in a population-based case-control study in a rural part of northern China during 2002-16.The authors found that women exposed to SHS were more likely to have a pregnancy affected by an OFC (all cases combined), and specifically with cleft lip with or without cleft palate (CL ± P) (OR 1.3 [95%CI 1.2, 2.2]). When the data were adjusted for maternal fever/ flu, infant sex, a history of a pregnancy previously affected by a congenital anomaly and farming occupation, the association remained the same. Importantly from a biological plausibility perspective, this study considered whether there was a dose-response effect; frequent SHS exposure, defined as more than 6 times per week, was associated with a higher risk of OFCs (OR 2.6 [1.8, 3.8]) and CL ± P (OR 2.5 [1.7, 3.7]) than exposure to SHS 1-6 times a week. 6Studies investigating risk factors for congenital anomalies are challenging as these are rare events which are often poorly recorded.The study by Pi and colleagues 6 has several stre...