The aim of this study was to evaluate the prevalence and the causing factors of job dissatisfaction in three subgroups of nurses working in different context: i) hospital wards for acute patients, ii) units for long-term patients, and iii) home care for chronic patients. An anonymous, self report questionnaire was used, composed by socio-demographic, burnout, stress and psychosocial (INRS) standardized variables. A specific summarizing item was also employed to evaluate the perception of job satisfaction. One hundred thirty seven out of 190 delivered questionnaires were returned (72.1%). About 30% of the total sample is unsatisfied, 28% of the sample suffered emotional exhaustion, 20% depersonalization, and 43% declared a low level of personal accomplishment. Stress scores beyond the threshold value of 17 were found for anguish (22.7 +/- 155), anxiety (22.3 +/- 15.8) and gastroenteric symptoms (19.0 +/- 17.8). Psychosocial scores beyond the threshold value of 50 were found for the variables general job load (56.7 +/- 19.8), attention (83.8 +/- 20.9), and social support from the supervisor (51.3 +/- 25.3). The logistic regression analysis suggested that coming from extracommunitarian Countries (OR: 3.76; CI: 1.04-13.65; p < 0.05), working with acute patients (OR: 2.94; CI: 0.85-10.11; p = 0.08), home-care of chronic patients (OR: 73; CI: 11.27-473.0; p < 0.001), assumption of psychodrugs (OR: 5.01; CI: 1.40-17.92 p < 0.01), and anxiety (OR: 3,71; CI: 1.20-11.44; p < 0.05), were factors significantly associated with a low job satisfaction compared to the other subgroups. On the contrary, a high work degree of work involvement (OR: 0.37; CI: 0.14-0.97; p < 0.05) resulted to be a protective factor respect to job disaffection. These findings suggest that changes that have been occurring in the last years in the social conditions and in health organization in Italy can deeply influence the degree of job satisfaction among the nurses. In order to improve the analysis of the welfare state of the health workers, more effective methods of analysis should be used, in addition to the questionnaires.
