Background In Tunisia, the information system on medical causes of death is based on the use of standard death certificate model based on international model recommended by the World Organization started in January 2001. However, this system is still burdened with a major death causes of under-registration. Only specific surveys on mortality have FAR generate reliable mortality indicators. Objectives to study the use of verbal autopsy in order to assess cancer mortality among Tunisian women in reproductive age (WRA) . Methods A retrospective national RAMOS survey (Reproductive Age Mortality Study). This survey was conducted in 2010 and included all deaths of women aged 15-49 years which occurred in 2008 and were due to cancer. Data were collected from civil status records and information gathered from families and from health institution's registers. For all deaths of women aged 15-49 years, the detailed circumstances and the sequence of events leading to death were grouped on a folder called "clinical record of verbal autopsy" .Then; all folders were submitted to the independent expert. Results During the study period, 1729 deaths among women of reproductive age (WRA) were the subject of a verbal autopsy against only 708 recorded by the National death information system (NDIS). Cancer is the leading cause of death among WRA .The specific rate of cancer mortality is 17.83 per 100 000 WRA against only 7.91 per 100 000 WRA estimated by the NDIS. Breast cancer is the leading cause with 35% of all cancers and specific death rate of 6.3 per 100,000 WRA against 2.48 per 100,000 WRA recorded by the NDIS. Conclusion Verbal autopsies Verbal autopsy remains an interesting method for measuring cancer mortality in women of a reproductive age especially in countries with a defective national death information system.