Aims and hypothesis: This review aimed to improve infant risk assessments in the context of maternal mental illness by identifying key predictors of poor parenting outcomes. Background: Inadequate parenting as a result of severe and persistent mental illness is a common reason for courts terminating parental rights. However, the current practice of parenting capacity assessments in the setting of perinatal psychiatry is fraught with risks and uncertainty. A well-recognised flaw in the assessment process is the lack of valid and reliable tools that have been specifically validated for assessing parenting capacity in mothers with a history of mental illness and the potential risk of harm to their infant. To date, there is only one instrument available. Methods: A systematic search of Medline, PsycInfo and Embase via the Ovid interface was conducted between September and December 2014. Citation snowball sampling was also used to identify further relevant studies. An additional search was performed in Google to access grey literature. Results: A total of 38 citations were identified, of which 8 publications focusing on the populations of England, France and Belgium met the eligibility criteria of this review. Evidence from existing research suggests that poor parenting outcomes in maternal psychiatric illness are strongly associated with correlates of socio-economic inequalities. However, evidence regarding the longterm implications of such factors is weak as only one follow up study and no longitudinal studies were identified in this review. Conclusion: Our review suggests that the use of standardised empirically validated risk assessment tools would benefit the current practice of parenting assessments by improving the process by which collected information is analysed. This would enhance the accuracy of decision-making, and improve the safeguarding of the infant. Further research is needed on medium to long-term parenting outcomes, particularly regarding its relations to: the type of maternal psychiatric disorder; the quality of maternal relationships; previous attachment experience; psychiatric illness or behavioural disorder in the partner and neonatal/infant medical complications. This would more accurately reflect the dynamic nature of parenting and would help to determine the effectiveness of specific interventions addressing risk factors associated with poor parenting outcomes.