Over the last years, labour markets are quickly changing due to ageing population, technological innovations and arising of new professions. In this evolving context, how can we manage â€œlongevity riskâ€? Is survival always increasing? Which are the variables that most affect the level of mortality? Our case of study focuses on disabled people, with regard to workers insured by Inail (National Institute for insurance against accidents at work), who suffered an accident at work or contracted an occupational disease and receive a life annuity. The results show that, in last decades, there was a general improvement of disabledâ€™ life expectancy as for Italian population, but for some groups of disabled it is not true. For example, we cannot expect any survival improvement for ill people with most serious diseases, such as all forms of cancer (including those related with asbestos). The variables that most affect the mortality of disabled are duration (years from the starting date of annuity to valuation date), severity of impairment (class of degree), type of event (accident or occupational disease) and disabledâ€™ profession or sector of economic activity (industry, agriculture, services). In case of high duration, because of the stabilization of eventâ€™s after-effects, mortality of disabled is wholly similar to Italian population one, and the other studied variables influence slightly the level of mortality. Otherwise, for lower duration, the other analyzed variables have a significant weight. There is a substantial difference between life expectancy of injured and occupational diseased: for low-mid impairment degrees, an injured survives about 10 years more than a diseased, for high impairment degrees over 20 years. In addition, our recent survey establishes that the sector of economic activity in which the disabled works is relevant: for example, an agricultural worker survives on average 5 years more than an industrial sectorâ€™s one.