to accept before requesting a CS with those of midwives and obstetricians and found that women were willing to accept higher risks than the professionals.[130] Women who do request CS report that it provides them with a sense of control over the birth and diminishes feelings of fear.[5,178] According to a study by Wagner et al. (2016), women’s fear is amplified by the risk-focused discourse in MNC. This generates the need for “risk talk”, contributing to a culture of risk, in which women are inclined to choose medical interventions. This, in turn, may give professionals the impression of women wanting interventions, leading to more interventions.[47,177] Although overuse in care is a sign of ineffective risk selection,[10,179] this outcome is not systematically evaluated and considered as a safety indicator. Instead, risk selection is focused on preventing underuse of care. And ostensibly this approach has increased the quality of MNC. Perinatal audits, for example, enable systematic assessment of practice against a defined standard, guided by the question ‘Could we have done more?’ recommendations aremade to address underuseof care.[180] This focus is understandable, considering that it is easier to determine the lack of medical interventions than the overuse of interventions; it can never be known for certain how not using an intervention would have affected the outcome of care. However, an increasing number of studies highlight the problem of the overuse of care, especially in high-income countries. Our study into the association between maternal characteristics and foetal and neonatal mortality reveals how a pathology informed notion of risk contributes to overuse and underuse of care and to care disparities. Focusing only on preventing underuse of care prohibits reflection on the use of unnecessary medical interventions and on the risks of overuse to maternal, foetal and neonatal health. It also contributes to a belief among care professionals that they are accountable for all outcomes experienced in pregnancy and birth, despite the legal position that healthcare professionals are responsible only for adverse outcomes caused by their own negligent actions.[139] This amplifies fear and anxiety amongst care professionals, increasing risk aversion and defensive practices,[60,181] including unnecessary referrals[182] due to fear of litigation.[5,25,182,183] From this perspective, risk emerges not from the pregnant and birthing body, but from MNC framing and managing pregnancy and birth as risky in response to pressures to account for MNC failure. The current conception of risk selection, however, may mask dysfunctional MNC practices. Thus, the notion of safe MNC can give rise to institutional risk itself.[111] Prevention before intervention While care is often used “too much too soon”,[3] care disparities exist, driven by societal determinants of health and mediated by institutional discrimination and inequity.[84,184– 189] Social determinants of health can be defined as conditions in the places where people 169 7 GENERAL DISCUSSION