Aims and objectives To compare the reliability and predictive validity of the Braden and Jackson/Cubbin PI risk assessment scales in intensive care unit patients. Background Risk assessment with a standardised tool is the usual intervention for preventing pressure injury. Therefore, tools used to assess pressure injury risk should be valid and reliable for the designated patient population. Design A prospective and cross-sectional study adheres to the STARD guideline. Methods This study was conducted between November 2017-April 2018 in the intensive care units of a tertiary level university hospital in Turkey. The study sample consisted of 176 patients admitted to three intensive care units. Risk assessment was performed once daily with the Braden scale, followed immediately with the Jackson/Cubbin scale. Risk assessment was terminated on the day of pressure injury development or upon patient discharge from the intensive care unit. Each patient's final risk assessment was considered in the data analysis. Results The Cronbach's alpha coefficient of the Jackson/Cubbin and Braden scales was .78 and .85, respectively. The predictive validity of the Jackson/Cubbin scale was confirmed by a sensitivity of .87, specificity of .84, positive predictive value of .47 and negative predictive value of .97. These values for the Braden scale were .95, .75, .38 and .99, respectively. Conclusion Both the Jackson/Cubbin and Braden scales are reliable and valid scales for pressure injury risk assessment in intensive care unit patients. However, the predictive ability to determine patients at risk and not at risk for pressure injury was better for the Jackson/Cubbin scale than for the Braden scale. Relevance to clinical practice Both scales are reliable and valid scales for pressure injury risk assessment. Jackson/Cubbin scale's discriminative ability (between the patients at pressure injury risk and not at pressure injury risk) was better.