Introduction: According to several studies, sick-leave during pregnancy plays an under-estimated role. It brings to light the existence of risk factors relating to strenuous physical work for contractions, sick-leave, hospitalizations, and premature births. The data obtained for all female workers at the AP-HP allow us to avoid all possible bias and to determine the potential impact of a policy of reducing strenuous physical work. Method: we were able to analyze data for a period of four years . The database used was PILOT RH, an intranet application providing information extracted from GIPSIE, which until 2008 was the institutional software of management of the careers of all the workers at the AP-HP. Results: In 2008, 3,937 women who had taken maternity leave (ML) accumulated 132,360 days of ordinary sick-leave (OL) (that is 33.6 days on average) without including the fifteen days allowed for “pathological pregnancy “(PP) and “post pregnancy leave.” Ordinary sick-leave for all the 53,132 women amounted to 11.5 days per person. But setting aside the women who took ML reduces the number of days of sick-leave to 9.7 days This difference accounts for 15.6 % of the total of OL absenteeism. The results are similar over the four years. Analysis by occupational category, for 2008, shows that the average number of days of sick-leave, after excluding women who took ML, decreases with the amount of strenuous work their post requires, as estimated by the PRESST-NEXT study: ancillary staff 19.9 days, nursing assistants 13.7 days, registered nurses 8.2 days, specialized nurses 7.7 days, medical-technical workers 5.7 days and head nurses 5.1 days. For the women who took ML, the average number of days of ordinary sick-leave, was highest for nursing assistants (41.9 days) followed by ancillary staff (36 days) and registered nurses (36 days), and was slightly less for specialized nurses (28.3 days). On the other hand, in the professions involving the least amount of physical labor, women take fewer than 25 days of ordinary sick-leave in the year of their pregnancy: medical-technical workers 21.3 days, head nurses 23.1 days and administration staff 23.7 days. The results are similar over the four years. For nurses, 25.5 % of total ordinary sick-leave is attributable to the women who had a pregnancy, rising to 34.4 % if we add OL and 15 days of official “pathological pregnancy.” Discussion: An intervention to reduce strenuous work in the test hospitals could be carried out. The reduction of sick-leave during pregnancy can be a quick indicator of evaluation of the efficiency of the investments. The improvements will benefit all the agents in the medium term, with a reduction in the risk of musculoskeletal disorders. It would also enable the employability of older employees to be maintained, as well as integrating people with a disability into a job that means a lot to them and that they do not want to give up.