A/HRC/42/44 Number and prevalence of persons in modern slavery, by category and region Region Number of Prevalence persons in (per thousand) forced labour of forced (in thousands) labour Number of persons in Prevalence forced (per thousand) marriage (in of forced thousands) marriage Number of Prevalence persons in (per modern thousand) of slavery (in modern thousands) slavery Africa 3 420 2.8 5 820 4.8 9 240 7.6 Americas 1 280 1.3 670 0.7 1 950 1.9 350 2.2 170 1.1 520 3.3 Asia and the Pacific 16 550 4.0 8 440 2.0 24 990 6.1 Europe and Central Asia 3 250 3.6 340 0.4 3 590 3.9 Arab States Source: ILO, Walk Free Foundation and IOM, Global Estimates of Modern Slavery: Forced Labour and Forced Marriage (Geneva, 2017). 5. One in four victims of contemporary forms of slavery in 2016 was thought to be a child. Women and girls were disproportionately affected, with over 71 per cent of victims being female. Of 24.9 million people experiencing forced labour, 4.1 million people were subject to State-imposed forced labour, 4.8 million people experienced forced sexual exploitation of adults and/or commercial sexual exploitation of children and 16 million people experienced forced labour exploitation in the private economy. The latter occurred in numerous settings, including domestic work (24 per cent), construction (18 per cent), manufacturing (15 per cent) and agriculture, forestry and fishing (11 per cent). Most victims were subjected to multiple forms of coercion at both the recruitment and employment stages. Almost 24 per cent of the 16 million victims of forced labour exploitation had their wages withheld, 17 per cent were under the threat of violence and 16.4 per cent of victims experienced physical violence. While physical violence affected men and women equally, 98 per cent of women and girls experienced sexual violence. 4 6. The understanding of where and why contemporary forms of slavery manifest is also improving, through the use of new statistical methods such as sentinel surveillance, network scale-up methods and multiple systems estimation.5 The latter technique has been used to estimate national prevalence in Ireland, the Netherlands, Romania, Serbia and the United Kingdom of Great Britain and Northern Ireland. And recent research and modelling have begun to identify vulnerability and risk factors, which can include age, gender, income, employment status, education level, health and other factors relating to social isolation. 6 Not all risks are linear: for example, young people are at higher risk than both children and older adults.7 4 5 6 7 Ibid. Walk Free submission, pp. 6���7; on sentinel surveillance, see United Nations University Institute on Computing and Society (UNU-CS) submission, Q6. See also United Nations Inter-Agency Project on Human Trafficking, Human Trafficking Sentinel Surveillance: Poipet, Cambodia, 2009–2010 (Bangkok, 2010) and Human Trafficking Sentinel Surveillance: Viet Nam-China Border 2010 (Bangkok, 2011). Pablo Diego-Rossel and Jacqueline Joudo Larsen, “Modelling the risk of modern slavery”, SSRN, 17 July 2018. See also Government of Malta and Walk Free submissions. Diego-Rossel and Larsen, “Modelling the risk of modern slavery”. For a child labour risk identification model in Latin America, developed in collaboration with the Economic Commission for Latin America and the Caribbean and ILO, see the website of the Latin America and the Caribbean Free of Child Labour Regional Initiative, http://iniciativa2025alc.org/child-labour-risk-model/. 3