EFSA received a request from the Hellenic Food Authority (EFET) for a scientific opinion on the risk to human health from the presence of nickel (Ni) in food, particularly in vegetables. The EFSA Panel on Contaminants in the Food Chain (CONTAM Panel) decided to extend the risk assessment also to drinking water. The reproductive and developmental toxicity in experimental animals was selected as the critical effect for the assessment of chronic effects of Ni. A tolerable daily intake of 2.8 µg Ni/kg body weight (b.w.) per day was derived from a lower 95 % confidence limit for a benchmark dose at 10 % extra risk (BMDL 10 ) of 0.28 mg/kg b.w. for post-implantation fetal loss in rats. The current dietary exposure to Ni raises concern when considering the mean and 95th percentile chronic exposure levels for all different age groups. The systemic contact dermatitis (SCD) elicited in Ni-sensitive humans after oral exposure to Ni was selected as the critical effect suitable for the assessment of acute effects of Ni. A lowest BMDL 10 of 1.1 µg Ni/kg b.w. was derived for the incidence of SCD following oral exposure to Ni of human volunteers. The CONTAM Panel applied a margin of exposure (MOE) approach and considered an MOE of 10 to be indicative of a low health concern. The MOEs calculated considering the estimated mean and the 95th percentile acute exposure levels were considerably below 10 for all age groups. Overall, the CONTAM Panel concluded that, at the current levels of acute dietary exposure to Ni, there is a concern that Ni-sensitized individuals may develop eczematous flare-up skin reactions. The CONTAM Panel noted the need for mechanistic studies to assess the human relevance of the effects on reproduction and development observed in experimental animals and for additional studies on human absorption of nickel from food, for example in combination with duplicate diet studies. Following a call for data on Ni levels in food and drinking water (water intended for human consumption and mineral waters), a total of 18 885 food samples and 25 700 drinking water samples were available in the final dataset to estimate dietary exposure to nickel. No speciation data were provided. Samples were collected between 2003 and 2012 in 15 different European countries, with almost 80 % of the total collected in one Member State. The most reported analytical methods were inductively coupled plasma-mass spectrometry (ICP-MS) and atomic absorption spectrometry (AAS), that represented 54 % and 42 % of the methods reported, respectively. The highest sensitivity was reported for the analysis of drinking water with a limit of quantification (LOQ) of 0.001 µg/L (for both ICP-MS and AAS). In food, ICP-MS showed the lowest LOQ for the analysis of 'Alcoholic beverages' (0.002 µg/kg) while the lowest LOQ reported with AAS was 1 µg/kg for samples of 'Fish and seafood' and 'Sugar and confectionery'. In the final dataset, left-censored data represented 66 % of the analytical results, with 35 % in food samples and 89 % in drinking water sa...