Background Mauritius embraces principles of a welfare state with free health care at point of use in any public facilities. However, the health financing landscape changed in 2007 when Private Health Expenditure (PvtHE) surpassed General Government Health Expenditure. PvtHE is predominately out of pocket (OOP) with only 3.4% related to premiums for private insurance. In 2014, Household OOP Expenditure on health accounted for 52.8% of total health expenditure. OOP is known to be regressive and to impact negatively on households’ living standards. Objectives This paper aims to examine trends in OOP in Mauritius, to assess its impacts through an analysis of key indicators of financial protection, namely catastrophic health expenditure (CHE) and impoverishment due to OOP health expenditure. It also aims to predict core determinants of CHEs. Methods Household Budget Surveys (HBS) of 2001/2002, 2006/2007 and 2012 were the primary source data. CHE and impoverishment were used to assess financial hardships resulting from OOP health payments. The incidence of CHE was estimated at three threshold levels (10,25 and 40%), using the budget share and the capacity to pay approaches. Impoverishment due to OOP was measured by changes in the incidence of poverty and intensity of poverty using the US$ 3.1 international poverty line. Logistic regression analysis was used to identify determinants of CHE. Findings Household CHE increased from 5.78% in 2001/02 to 8.85% in 2012 and 0.61% in 2001/02 to 1.25% in 2012, for 10 and 40% thresholds, respectively. The incidence of CHE was significantly higher in urban areas compared to rural areas. The highest levels of CHEs were among households’ heads, who are retired rising from 1.62% in 2001/02 to 3.71% in 2012, followed by households’ head who are widowed from 2.29% in 2001/02 to 2.63% in 2012 and homemakers from 2.12% in 2001/02 to 2.57% in 2012 at the 40% threshold. The share of households pushed below the poverty line due to OOP dropped from 0.4% in 2001/02 to 0.2% in 2006/07 before rising to 0.34% in 2012. In 2012, poverty gap occurred only among households under poorest quintile 1 (0.24%) and quintile 2 (0.03%). Overall poverty gap dropped from 0.08% in 2001/02 to 0.05% in 2012. Logistic regression analysis revealed that the odds ratio of facing CHE were significant only among households with heads being retired and with a presence of an elderly member in the household. Conclusion Despite the rise in incidence of CHE between 2001 and 2012 the impact of OOP on the level of impoverishment and poverty gap has not been significant.