Background and aim: Earlier research has suggested that local communities develop different sickness absence practices, expressed in variation in the absence figures. This study aimed to test the hypothesis that the work places residing in a locality share similar sickness absence practice. Methods: Register data from the major public sector employers and from private industrial employers were used to compare the absence in two localities. Results: There were significant differences by locality in both branches and both in short and in long sickness absences. Population-level statistics showed a corresponding locality difference in recipients of sickness allowance benefits and in primary healthcare visits, but not in indicators of mortality and morbidity. Social problems were more common in the locality with high rate of absence. Conclusion: Area-specific practice, or habitus, as an independent determinant of sickness absence is discussed and an explicitly sociological context is recommended for debates about "best sickness absence practices" and corresponding intervention measures.